The Impact of Morality and Prosociality on Power:
How Immoral People Can Gain Power
28
thJune 2017
Internship Report
Annika Nieper (11391634)
Supervisor: Prof. Dr. Gerben van Kleef
University of Amsterdam
Abstract
Many researchers have examined the impact of power on morality and there is a general consent that power corrupts. However, the reversed relationship has so far been mostly neglected: the possibility that immoral behavior empowers. Here, we examined the hypotheses that moral violators can gain power in the eyes of others and are also afforded power, as long as their actions are prosocial. Two studies, using different manipulations of prosociality, mostly support those ideas. In the first study, the a priori intentions were alternated, whereas in the second study the focus was on the a posteriori consequences of an action. In both studies, prosociality indeed had a positive impact on power affordance and especially prosocial intentions enabled moral violators to also gain power. Additional analyses showed that this effect was mediated by perceived social engagement. For power perception, we also found an impact of behavior as predicted, however, not consistently on two different scales and prosociality had a similar impact. Discrepancies between the two scales and
implications for power affordance, morality and social engagement are discussed.
The Impact of Morality and Prosociality on Power: How Immoral People Can Gain Power
Newspaper articles are filled with examples of powerful people engaging in immoral actions. Those led many people conclude that power corrupts, which is supported by scientific research (e.g. Kipnis, 1972). However, what happens when a person engages in an immoral action? Do they lose their power or does this act even advance their power position? While one would hope perhaps that people, who engage in immoral actions, will fall from grace and lose their power, many politicians suggest otherwise. For example, Donald Trump, who has shown during his campaign repeatedly immoral behavior, was still able to win the elections. Similarly, Joschka Fischer, who was famously known in Germany for physically attacking police officers, was later elected as Vice Chancellor in 1998. Thus, what is the impact of morality on power? Can moral violators gain power in the eyes of others? And further, when are they also afforded power? We explored those possibilities by focusing on the impact of moral violations and prosociality on perceptions and affordance of power.
What are moral violations?
Norms are principles or rules that are commonly accepted within a group and guide and/or constrain behavior (Cialdini & Trost, 1998). Those can be social norms (e.g. interrupting others) as well as moral norms (e.g. harming others; Brennan, Eriksson, Goodin, & Southwood, 2013). Accordingly, moral violations are behaviors that infringe the latter. What do people consider as moral norms? According to the Moral Foundation Theory (Graham et al., 2013), people generally consider five
foundations (care/harm, fairness/cheating, loyalty/betrayal, authority/subversion, sanctity/ degradation) as morally important. While for example liberals do not regard the violation of the last three foundations as immoral (Graham, Haidt, Nosek, & Carver, 2009), the care/harm foundation is considered as morally important across cultures (Haidt, 2008) and is independent of political affiliation. This research focuses on the care/harm foundation of the Moral Foundation Theory and examines the possibility that moral violators are perceived as more powerful. Further, they are also afforded power as long as the immoral behavior is intended to benefit the society (Study 1) or the consequences are beneficial for the society (Study 2).
Immoral behavior can signal power
Powerful people are more likely to show or are believed to show certain behaviors and those behaviors themselves can signal power. According to the approach/inhibition theory of power (Keltner, Gruenfeld, Anderson, 2003), power disinhibits behavior. High power individuals feel fewer social constraints and will more likely engage in actions that satisfy their current needs. In comparison, low power individuals feel constrained by potential punishments and therefore restrict their actions. Thus, as power disinhibits behavior, powerful individuals are more likely to engage in immoral actions. Indeed, studies have shown that powerful
individuals are more likely to betray their romantic partner (Lammers, Stoker, Jordan & Pollmann, 2011), ignore the suffering of others (Van Kleef et al., 2008), treat others as a means to their own ends (Gruenfeld, Inesi, Magee & Galinsky, 2008) and endorse immoral behaviors themselves while imposing stricter morals on others (Lammers, Stapel & Galinsky, 2010). As people have rich stereotypes of the behaviors of
powerful people (Hall, Coats & LeBeau, 2005; Keltner, Van Kleef, Chen & Kraus, 2008), the immoral behavior by itself can be a cue for the observer and can signal power.
Furthermore, immoral behavior may not only be associated with power, it also shows that the violator can act of his own volition. As power is defined as the asymmetric control over valued resources in social relations (Magee & Galinsky, 2008), powerful individuals have more control and are therefore less dependent on others. This lack of dependency reduces the pressure to conform to norms (Galinsky, Magee, Gruenfeld, Whitson & Lijenquist, 2008) and people are free to act the way they want. People indeed associate being free of constraints and behaving the way you wish with power. For example, people who display greater action orientation are perceived as more powerful because they can act according to their own volition (Magee, 2009). Similarly, social norm violators are perceived as more powerful due to their apparent autonomy and free will (Van Kleef, Homan, Finkenauer, Gündemir, & Stamkou, 2011). By the same logic, violating moral norms suggests that the
violator has high levels of power that enables him to behave the way he pleases. Thus, moral violators can be perceived as more powerful because the immoral behavior itself can signal power and suggests high volitional capacity, a freedom associated with power (Hypothesis 1).
The importance of prosociality for power affordance
While often power perceptions and power affordance go hand in hand, they can be disconnected. Many people for instance perceive Donald Trump as powerful, but are simultaneously not willing to grant him more power by for example voting for him. When are people additionally afforded more power? The reciprocal influence model of power (Keltner et al., 2008) suggests that people are afforded power if they advance the interests of the group. Thus, people who show social behaviors that benefit others or the society as a whole are afforded with power. This also holds true for norm violators (Van Kleef, Homan, Finkenauer, Blaker, & Heerdink, 2012). In the experiment, a confederate stole a cup of coffee and offered it to the participant. Even though he violated a social norm, the confederate was afforded more power.
Interestingly, the confederate that offered a cup of coffee to the participant without violating the norm (it was allowed) was not afforded more power. Perceived social engagement mediated this effect. Thus, prosocial norm violators were perceived as more socially engaged and were afforded more power in comparison to people who behaved prosocially without violating the norm. Although this study indicates that people, who violate social norms, can indeed gain power if they are prosocial, is this also the case for moral violators? Specifically, are people, who physically harm others in order to advance the interest of the group, really still afforded power? While one would hope perhaps that physically harming others is so severe that people do not want such as person as their boss, we hypothesize that moral violators, even by physically harming others, can also gain power, as long as they are prosocial (Hypothesis 2).
What is prosocial behavior?
Some definitions of prosocial behavior place the emphasis on the consequences of an action and refer to behaviors that are beneficial to others such as helping (e.g. Stürmer & Snyder, 2010). On the other hand, other definitions stress the importance of the intention underlying a certain behavior (e.g. Batson & Powell, 2003) and prosocial behavior is behavior that intends to help others. Therefore, both - intentions and consequences – seem to be important factors that people might consider when judging prosociality. Importantly, by definition prosocial behavior is not limited to “nice” behavior, it can also be punitive or aggressive that is nonetheless beneficial for others. For instance, punishing free riders can be beneficial for others and increase cooperation (Fehr & Gächter, 2002). In contrast, people whose behaviors are only beneficial to them or intended to only benefit themselves, are often referred to as selfish.
The current research
The goal of the study is to examine whether moral violators are perceived as more powerful and when they are also afforded more power. As immoral behavior signals power, it is hypothesized that moral violators are perceived as more powerful as opposed to people who obey the rules. Further, moral violators can also gain power if they are prosocial. As the definitions of prosociality suggest, intentions as well as consequences qualify to make an action prosocial. Thus, in the first study we will manipulate prosociality with the a priori intentions of the actor, whereas in the second study, we alternate the a posteriori consequences of the action.
Study 1
Method
Two-hundred-four participants (136 females; mean age = 28.96, SD = 10.84) were randomly assigned to the conditions of a 2 (moral violation vs. control) x 2 (prosocial vs. selfish) design. They were presented with a scenario, which was orientated on a moral dilemma used by Greene, Nystrom, Engell, Darley & Cohen (2004). In this dilemma, passengers of a cruise ship have to abandon the ship due to a fire. The lifeboats are carrying many more people than they were designed to carry and something has to be thrown overboard for the passengers to survive. In the immoral condition, the main character decides to push an injured person overboard, who will not survive in any case. In comparison, in the non-violating condition, the main character decides to throw bags overboard. In this study, prosociality was manipulated with the intentions of the actor. Prior the actor engages in the action, information about his intentions are revealed: “(the actor) intends to save as many people as possible (vs. himself)”. To emphasize the manipulation, we also added at the end “… with the hope of saving as many people as possible (vs. himself).
Materials
We were interested in participants’ perceptions of the actor’s power and whether he would also be afforded power. We used two scales to measure power perceptions. The first one is the validated Generalized Sense of Power scale by Anderson, John and Keltner (2012). An example item is: “In his relationship with others, his ideas and opinions are often ignored” (reversed, α = .77). In addition,
participants were asked to rate the main actor on six adjectives indicating power such as decisive, strong and powerful (α = .64). For power affordance, participants rated the extent they would like the main actor as their boss, as their political leader and whether they would give him a promotion (α = .90). Furthermore, the Leadership Endorsement scale (Stamkou et al., 2016) was used. For this scale, participants imagined that the main character was running for a leadership position within the same company as they are working in. Then, they had to indicate the extent they would support him. Example items are “I would vote for ___” and “I would be a strong supporter of ___” (α = .88). For exploratory purposes, we added perceived social engagement (Keltner et al., 2008) and participants indicated how social,
assertive, helpful, sympathetic and friendly they found the main actor (α = .84). Those questions were embedded in a questionnaire on “social perceptions” to conceal the study’s purpose1 and all items were measured on a scale from 1 to 7 (definitely not to definitely yes). We used factual questions to check whether participants understood the situation. Participants had to indicate the extent the main actor intended to save as many as possible and furthermore, the extent to which he actively harmed somebody else.
Results
Manipulation Check. Participants in the immoral condition properly
indicated that the main actor harmed somebody else (M = 5.42 SD = 1.79) in comparison to the moral condition (M = 2.24, SD = 1.47, F(1, 200) = 193.57, p < .001, !2 = 0.49), indicating that the manipulation was effective. No other effects
1 In this questionnaire, we also explored the impact of morality and prosociality on moral emotions, volition
inferences, perceived morality (behavior), competence, warmth and morality (character). Those results can be found in Appendix A.
occurred (both Fs < 2.25, ns). Further, participants indicated in the prosocial
condition that the main actor had the intention to save as many as possible (M = 5.91, SD = 1.68) in comparison to the selfish condition (M = 3.21, SD = 2.06), F(1, 200) = 106.14, p < .001, !2 = 0.34. There was no effect of behavior, F < 1, but a marginal significant interaction occurred, F(1, 200) = 13.34, p = .054, !2 = 0.01. Simple effects revealed that within the prosocial condition, intentions did not significantly differ between the moral (M = 6.18, SD = 1.47) and immoral condition (M = 5.65, SD = 1.85), F(1, 200) = 2.03, p = .16. Similarly, within the selfish condition, intentions did not significantly differ between the moral (M = 2.96, SD = 1.99) and immoral
condition (M = 3.45, SD = 2.13), F(1, 200) = 1.74, p =.19. A few participants made mistakes on the manipulation check, however, due to the large sample size, we didn’t exclude any participants. A non-significant fisher exact test indicated that the few mistakes2 were rather random on both manipulation checks (pMP1 = 0.09 and pMP2 =
0.74).
Power perception. Power perception was analyzed using a 2 (Behavior:
immoral vs. moral) x 2 (Intentions: prosocial vs. selfish) mixed ANOVA. A main effect of behavior indicated that power perception was higher in the immoral (M = 4.98, SD = 0.85) than in the moral condition (M = 4.74, SD = 0.83), F(1, 200) = 4.43, p = .037, !2 = 0.02. In addition, a main effect of intentions indicated that power perception was higher in the prosocial (M = 4.99, SD = 0.86) than in the selfish condition (M = 4.73, SD = 0.81), F(1, 200) = 4.88, p = .028, !2 = 0.02. No interaction emerged, F < 1, as can be seen in Figure 1. A post-hoc Tukey test further revealed that the immoral and prosocial violator (M = 5.13, SD = 0.92) was perceived as significantly more powerful than the moral and selfish person (M = 4.62, SD = 0.87, p
2 Ratings lower than 4 in the immoral (or prosocial) conditions and higher than 4 in the moral (or selfish)
= .014). All other comparisons were non-significant.
Figure 1: Impact of behavior and intentions on power perception
On the second measurement,a slightly different pattern occurred. There was a main effect of intentions F(1, 200) = 17.92 p < .001, !2 = 0.08, and no main effect of behavior, F < 1, but a marginal significant interaction F(1, 200) = 2.99, p = .085, !2 =
0.01. Within the prosocial condition, the immoral person was perceived as similar powerful (M = 5.34, SD = 0.85) as the moral person (M = 5.16, SD = 0.74), F(1, 200) = 1.15, p = .28. Similarly, in the selfish condition, the immoral person was perceived as similar powerful (M = 4.67, SD = 0.74) as the moral person (M = 4.87, SD = 0.81), F(1, 200) = 1.88, p = .17.
Power affordance. Furthermore, we analyzed power affordance using a 2
(Behavior: immoral vs. moral) x 2 (Intentions: prosocial vs. selfish) mixed ANOVA. For the first scale, a main effect of behavior indicated that power affordance was higher in the moral condition (M = 4.30, SD = 1.55) than in the immoral condition (M = 3.00, SD = 1.41), F(1, 200) = 44.44, p < .001, !2 = 0.16. In addition, a main effect
4.0 4.5 5.0 5.5 Immoral Moral Behavior P ower P ercept ion Intentions Prosocial Selfish
of intentions indicated that power affordance was higher in the prosocial (M = 4.17, SD = 1.37) than in the selfish condition (M = 3.13, SD = 1.68), F(1, 200) = 28.73, p < .001, !2 = 0.11. No interaction emerged, F < 1. The second measurement (leadership
endorsement scale) showed an identical pattern. Participants afforded more power to the moral characters (M = 4.66, SD = 1.28) than to the immoral characters (M = 3.73, SD = 1.28), F(1, 200) = 29.10, p < .001, !2 = 0.11. In addition, participants afforded
more power to the prosocial characters (M = 4.74, SD = 1.13) than to the selfish characters (M = 3.65, SD = 1.48), F(1, 200) = 39.69, p < .001, !2 = 0.16. No
interaction emerged, F < 1, as can be seen in Figure 2.
Figure 2: Impact of behavior and intentions on power affordance
Perceived Social Engagement. A similar pattern to power affordance was
found for perceived social engagement. Participants perceived the moral actors as more socially engaged (M = 4.82, SD = 1.18) in comparison to the immoral characters (M = 3.78, SD = 1.09), F(1, 200) = 52.37, p < .001, !2 = 0.18. Intentions also
impacted perceived social engagement and the prosocial characters were perceived as more socially engaged (M = 4.75, SD = 1.18) than the selfish characters (M = 3.85,
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 Immoral Moral Behavior P ower A ffordance Intentions Prosocial Selfish
SD = 1.15), F(1, 200) = 39.54, p < .001, !2 = 0.14.In addition, there was a small interaction of behavior with intentions, F(1, 200) = 4.70, p = .031, !2 = 0.02 and
simple effects revealed that within the immoral condition, the prosocial person was perceived as more socially engaged (M = 4.07, SD = 0.99), than the selfish person (M = 3.48, SD = 1.11), F(1, 200) = 8.49, p = .004, !!2 = 0.04. Even stronger was the contrast within the moral condition. The prosocial and moral person (M = 5.44, SD = 0.95) was perceived as a lot more socially engaged than the selfish moral person (M = 4.21, SD = 1.08), F(1, 200) = 35.76, p < .001, !!2 = 0.15.
Figure 3: Impact of behavior and intentions on perceived social engagement
Furthermore, we explored the possibility that the impact of behavior on power affordance was mediated by perceived social engagement. As behavior and intentions interact on the latter, we conducted a moderated mediation analysis (Model 7, Hayes, 2013) with intentions as a moderator. To test for a moderated mediation, we examined the indirect effect of perceived social engagement on the relation between behavior and power affordance on each level of the moderator. Using bootstrapping with 5000 samples, there was a significant indirect effect in the prosocial condition [.731,
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Immoral Moral Behavior P erceived S ocial E ngagement Intentions Prosocial Selfish
1.350], and in the selfish condition [.233, .907], as the 95% bias corrected confidence intervals did not contain zero. Importantly, a test of equality of the conditional
indirect effects in the two groups revealed that the 95% bias corrected confidence interval also did not contain zero [-.890, -.056], indicating that the two levels differ significantly and supporting moderated mediation. The direct effect of behavior on power affordance was reduced to non-significance (p = .36), suggesting full mediation by perceived social engagement.
Discussion
The first study gives initial insight into the power perception and affordance of moral violators. Specifically whether people, who violate moral norms, are perceived as more powerful and further whether they are also afforded power, as long as they have prosocial intentions. Results mostly confirmed our expectations.
First, in regards to the orthogonality of the manipulation checks, the second manipulation check was not completely orthogonal. Intentions slightly interacted with behavior. However, as the size of this effect was really small, this is considered to have only little implications on the interpretations of the results.
In regards to the first hypothesis, we indeed found an impact of behavior on power perception on the first scale as predicted, thus replicating previous findings that norm-violators are attributed more power than norm-abiders. Surprisingly, intentions had a similar effect on power perception and on the second scale we only found an impact of intentions. Further analysis revealed that the two scales did not correlate highly3 (r = .42), indicating low convergent validity and the possibility that the two
scales might measure slightly different constructs. Previous research already suggested that there is a difference between wanting power over others, sometimes referred to as social power, and power as a need for autonomy, sometimes referred to as personal power (Lammers, Stoker, Rink & Galinsky, 2016). Potentially the
Generalized Sense of Power4 scale rather assesses the perceptions of social power, the power one has over others, while the second scale with the powerful adjectives5 rather
assesses personal power, how free of the influence of others the main actor is or perceptions of power independent of social relationships. This could eventually explain the discrepancy between the findings on the two scales. Overall, on both scales, the prosocial and immoral person could gain power in the eyes of others and was perceived as more powerful than the moral and selfish person on the first scale and than the two selfish characters on the second scale.
The second hypothesis was confirmed: prosociality indeed enables moral violators to gain power and selfish vs. prosocial intentions had a large effect on power affordance, also highlighting the harming impact of selfish intentions. Exploratory analysis showed that the effect of behavior on power affordance was fully mediated by perceived social engagement and intentions attenuated this relationship through the link between behavior and perceived social engagement. Those findings are in line with the reciprocal influence model of power and in particular the social engagement hypothesis, that those that are perceived as more socially engaged are afforded more power.
4 Example items: „In his relationship with others, ____ can get others to do what he wants“ and „In his relationship
with others, if he wants to, he gets to make the decisions.
In sum, people can gain power in the eyes of others if they act immorally, however only if they are prosocial, indicating that they might not be perceived as more powerful regardless of their intentions. Further, moral violators can still gain power as long as they have prosocial intentions. Prosocial moral violators are perceived as more socially engaged in comparison to the selfish moral violators, enabling them to gain power. While the focus of the first study was on the a priori intentions, we examined whether the a posteriori consequences have a similar impact on power perception and affordance in the second study.
Study 2
Method
The design was the same as in study 1. Two-hundred-four psychology students (141 females; mean age = 20.65, SD = 2.20) participated in exchange for course credit and were randomly assigned to one of the four conditions. The scenarios were orientated on the same moral dilemma as in study 1 and the manipulation of morality was identical. The only difference is the manipulation of prosociality. In the first study, we manipulated prosociality with the intentions of the actor, whereas in this study, it is manipulated with the consequences of the action. In the prosocial condition, the lifeboat is overcrowded and something has to be thrown overboard for all the passengers to survive. The main character engages in an action (immoral vs. control) and at the end it said: “Through his action, … all passengers survive”, thus the beneficial consequences of the action are mentioned and highlighted. In
comparison, in the selfish condition, the main character is still on the sinking cruise ship and the lifeboats are already overcrowded. In order for him to board, he needs to
throw something overboard. The main actor engages in an action (immoral vs. control) and at the end it said: “Through his action, … (the actor) survives”, highlighting the fact that the action only had benefits for the main character.
Materials
The materials were identical with study 1. As a manipulation check, we used factual checks and participants indicated the extent the main character actively harmed somebody else and the extent he saved others through his action.
Results
Manipulation Check. Participants in the immoral condition properly
indicated that the main actor actively harmed somebody else (M = 5.52 SD = 1.65) in comparison to the moral condition (M = 2.75, SD = 1.68, F(1, 200 = 141.06, p < .001, !2 = 0.41), indicating that the manipulation was effective. No other effects occurred, both Fs < 1. Further, participants indicated in the prosocial conditions that the main actor saved others through his action (M = 6.13, SD = 1.12) in comparison to the selfish condition (M = 3.79, SD = 1.99), F(1, 200) = 105.74, p < .001, !2 = 0.34) and
no other effects occurred, both Fs < 1. Identical with the first study, we didn’t exclude any participants and a Fischer’s exact test indicated that the mistakes were rather random (pMP1 = .67 and pMP2 = .30).
Power perception. Power perception was analyzed using a 2 (Behavior: immoral
vs. moral) x 2 (Consequences: prosocial vs. selfish) mixed ANOVA. A main effect of behavior indicated that power perception was higher in the immoral (M = 5.21, SD = 0.73) than in the moral condition (M = 5.00, SD = 0.70), F(1, 200) = 4.57, p = .034, !2
significant interaction, F(1, 200) = 3.40, p = 0.07, !2 = 0.02. Simple effects revealed that within the prosocial condition, the immoral person was perceived as more powerful (M = 5.34, SD = 0.81) than the moral person (M = 4.94, SD = 0.73), F(1, 200) = 7.93, p = .005, !!2 = 0.04. Whereas within the selfish condition, there was no
difference between the immoral (M = 5.09, SD = 0.61), and the moral person (M = 5.06, SD = 0.68), F(1, 200) = .044, p = .83, as can be seen in Figure 4.
Figure 4: Impact of behavior and consequences on power perception
On the second measurement, a slightly different pattern occurred. There was a main effect of consequences F(1,200) = 4.30, p = .039, !2 = 0.02, and no main effect of behavior, F(1,200) = 2.69, p = .10, !2 = 0.01, but a marginal significant interaction, F(1,200) = 3.22, p = .07, !2 = 0.02. Simple-effects analysis revealed that the prosocial moral violator was perceived as more powerful (M = 5.31, SD = 0.79) in comparison to the selfish moral violator (M = 4.90, SD = 0.77), F(1, 200) = 7.47, p = .007, !!2 = 0.04. There was no difference in the moral condition between the prosocial (M = 5.29, SD = 0.67) and selfish person (M = 5.26, SD = 0.78), F(1, 200) = 0.39, p = .84.
4.0 4.5 5.0 5.5 Immoral Moral Behavior P ower P ercept ion Consequences Prosocial Selfish
Power affordance. Furthermore, we analyzed power affordance using a 2
(Behavior: immoral vs. moral) x 2 (Consequences: prosocial vs. selfish) mixed
ANOVA. For the first scale, a main effect of behavior indicated that power affordance was higher in the moral condition (M = 4.71, SD = 1.31) than in the immoral
condition (M = 3.32, SD = 1.07), F(1, 200) = 69.91, p < .001, !2 = 0.26. In addition, a
main effect of consequences indicated that power affordance was higher in the prosocial (M = 4.18, SD = 1.29) than in the selfish condition (M = 3.85, SD = 1.45), F(1, 200) = 4.01, p = .047, !2 = 0.01. No interaction emerged, F < 1. The second
measurement (leadership endorsement scale) showed an identical pattern. There was a main effect of behavior indicating that power affordance was higher in the moral condition (M = 5.05, SD = 0.95) than in the immoral condition (M = 3.71, SD = 1.23), F(1, 200) = 77.67, p < .001, !2 = 0.27. In addition, a main effect of consequences
indicated that power affordance was higher in the prosocial (M = 4.60, SD = 1.13) than in the selfish condition (M = 4.17, SD = 1.40), F(1, 200) = 8.01, p = 0.005, !2 =
0.03. No interaction emerged, F < 1, as can be seen in Figure 5.
Figure 5: Impact of behavior and consequences on power affordance
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 Immoral Moral Behavior P ower A ffordance Consequences Prosocial Selfish
Perceived social engagement. Highly similar to power affordance is the
pattern of perceived social engagement. Here we found an impact of behavior and participants perceived the characters in the moral conditions as more socially engaged (M = 5.01, SD = 0.92) than in the immoral condition (M = 3.73, SD = 0.92), F(1, 200) = 101.48, p < .001, !2 = 0.32. In addition, consequences had an impact and people in
the prosocial conditions were perceived as more socially engaged (M = 4.64, SD = 1.07) than in the selfish conditions (M = 4.10, SD = 1.15), F(1, 200) = 17.92, p < .001, !2 = 0.06. No interaction occurred, F < 1, as can be seen in Figure 6.
Figure 6: Impact of behavior and consequences on perceived social engagement
Furthermore, we explored the possibility that perceived social engagement mediates the impact of prosociality and behavior on power affordance. First, we tested for the impact of consequences and examined the indirect effect of perceived social engagement on the relation between consequences and power affordance. Five-thousand bootstrap samples revealed that the estimates of the indirect effect were negative and the 95% bias-correct confidence interval did not contain zero [-.644,
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 Immoral Moral Behavior P erceived S ocial E ngagement Consequences Prosocial Selfish
-.160]. The direct effect of consequences on power affordance was reduced (p = .77), suggesting full mediation. Furthermore, the impact of behavior on power affordance was partially mediated. The 95% bias-corrected confidence interval did not contain zero [.519, 1.013] and the direct effect was reduced, however still significant (p < .001), suggesting partial mediation.
Discussion
The second study investigated the impact of morality and consequences on power perception and affordance. Here, we found an impact of behavior on power perception on the first scale, however not on the second. Consequences interacted with behavior on both scales, indicating that moral violators are not attributed more power simply because they show immoral power, but that consequences also matter. This pattern of results is highly similar to the first study and also in this study, there was not as a high correlation between the two scales as one would expect if they would measured exactly the same construct (r = .56). This further supports the idea that the two scales might measure slightly different constructs where different mechanisms might play a role. Interesting however is that on both scales, the prosocial and immoral person could gain power in the eyes of others and was
perceived as more powerful than the prosocial and moral person on the first scale and than the immoral and selfish person on the second scale. In regards to power
affordance, prosocial consequences had a positive impact on power affordance and further exploratory analyses showed that perceived social engagement is a significant mediator between the impact of behavior and prosociality on power affordance.
General discussion
The idea that powerful people are more likely to engage in immoral actions is widespread (e.g. Kipnis, 1972, Lammers et al., 2010). However, whether immoral people can also rise to power has so far not been investigated. Although one would hope, perhaps, that people are sensitive to immoral behavior and moral violations undermine power, the present investigation revealed that this is not necessarily the case. Building on previous research on social norms, we argued that moral violators can gain power in the eyes of others and are also afforded power as long as they are prosocial. Both studies consistently showed that prosocial moral violators could gain power in the eyes of others and were perceived as more powerful than some other characters. This was however not the case for selfish moral violators, indicating that moral violators might not gain power in the eyes of others regardless of the social consequences. Furthermore, prosociality enabled moral violators to also gain power. Especially prosocial intentions impacted power affordance positively, interestingly even more than in comparison to consequences in the second study. Perceived social engagement emerged as a meaningful mediator and individuals who violate moral norms prosocially were perceived as more socially engaged in comparison to selfish moral violators, which enabled them to gain power.
Previous findings on social norms already suggested that norm violators are perceived as more powerful than individuals who obey the rules (Van Kleef et al., 2011) and further that they can also gain power as long as they are prosocial (Van Kleef et al., 2012). It was unclear however, whether this effect was limited to specific types or degrees of norm violations. The present studies extended those findings by investigating the impact of moral violations and indicate that social and moral norm
violations are similar yet distinctly different. Previously it was found that people who break social norms for the benefit of others were actually perceived as more socially engaged and afforded more power in comparison to people who obeyed to those rules (Van Kleef et al., 2012). This was not the case for moral violations and in both studies it was, perhaps reassuringly, the person that is moral and prosocial that was afforded most power. This was also suggested by Keltner (2016) who argues that the person that is most utilitarian and thus minimizes harm and maximizes the benefits is afforded most power. In the present studies, the moral violators caused harm and therefore it was the prosocial and moral person that was most utilitarian and also afforded most power.
Both studies lend support to the reciprocal influence model of power, especially their social engagement hypothesis (Keltner et al., 2008). Those that are perceived as more socially engaged are also afforded more power. While so far studies mostly focused on the consequences of an action, the present studies suggest that intentions might be another factor that highly influences perceived social engagement. Further, by giving support to the reciprocal influence model of power, those studies suggest that it is not as Machiavelli stated that one should be immoral and frightening towards others in order to maintain power (Machiavelli, 1975), but rather underline that people should be prosocial or perceived as socially engaged if they want to gain power.
This conclusion points to a mechanism on how immoral people can gain power and maintain leadership positions. Powerful people are more likely to engage in immoral actions (Kipnis, 1972; Lammers et al., 2010). When these moral violations
entail benefits for others or are intended to benefit others, moral violators can keep their leadership position and even gain more power, which keeps corrupt people in powerful positions. This undesired cycle might be witnessed for example in politics. Even though many politicians have shown corrupt behavior, they often did not fall from grace but maintained their position and sometimes even gained more power. Similarly in other areas, for example in banking, employees might not mind a
corruptive boss, as long as they themselves benefit from it through for example higher bonuses.
In this context, it is important to note that we are not trying to claim that people should violate moral norms prosocially in order to gain power, but we first demonstrated that people, even though they physically harmed somebody for the benefit of the group, can gain power. In both studies it was, perhaps reassuringly, the person that is moral and prosocial that was afforded most power. Furthermore, it is also unclear whether prosociality always enables moral violators to also gain power. The present studies focused on the care/harm foundation. Even though violations of this foundation are consistently considered as severe and immoral (Haidt, 2008), one should be careful with generalizing those finding to other moral violations. For instance, people might find loyalty or fairness in leaders highly important and a violation of those might result in no power affordance at all. Furthermore, the findings above also might vary for different types of harmful behavior. While the present studies concentrated on physical direct harm, it might be different for indirect or even economical harm. Future studies could extend those findings by investigating
Furthermore, it would also be interesting to look at individual differences in endorsing non-normative actions in leaders. Previous research already suggested that the observer’s verticality influences the extent norm violators rise to the top (Stamkou et al., 2016). One could further look into individual differences by for example
examining political affiliation. As Graham et al. (2009) suggested what people find morally important differs between liberals and conservatives. Going against an authority figure for example is something that is not considered as immoral by liberals. Could liberals maybe afford people who violate the authority foundation more power than conservatives? If the observer benefits from this violation, the moral violator might then be perceived as more socially engaged, potentially even more in comparison to somebody who does not violate this foundation. This might eventually lead to even more power affordance in comparison to somebody who is prosocial without violating the moral norm. Future studies could investigate those possibilities.
Lastly, the two present studies suggest that intentions might matter more than consequences. Future studies could specify this difference more. It could potentially be that this differs in behavioral settings. When somebody actually benefits from a specific behavior, this might impact power affordance more than intentions, whereas the latter might play a bigger role when reading about a person in a newspaper.
Potentially intentions allow more inferences about the character of a person and one is better able to predict future behavior as opposed to knowing the consequences of an action. Examining those questions will further increase knowledge on how people can gain power.
References
Anderson, C., John, O., & Keltner, D. (2012). The Personal Sense of Power. Journal
of Personality, 80(2), 313-344.
Batson, C. D. and Powell, A. A. 2003. Altruism and Prosocial Behavior. Handbook of Psychology. Three:19:463–484.
Brennan, G., Eriksson, L., Goodin, R. E., & Southwood, N. (2013). Explaining
norms. Oxford University Press.
Cialdini, R. B., and Trost, M. R. (1998). Social influence: social norms, con- formity, and compliance. In D. T. Gilbert, S. T. Fiske, and G. Lindzey (Eds.), Handbook of social psychology (4th edn., vol. II, pp. 151–92). Boston: McGraw-Hill. Fehr, E., & Gächter, S. (2002). Altruistic punishment in humans. Nature, 415(6868),
137-140.
Galinsky AD, Magee JC, Gruenfeld DH, Whitson JA, Liljenquist K (2008) Power reduces the press of the situation: implications for creativity, conformity, and dissonance. Journal of Personality and Social Psychology, 95:1450-1466. Graham, J., Haidt, J., Koleva, S., Motyl, M., Iyer, R., Wojcik, S., & Ditto, P. H.
(2013). Moral Foundations Theory: The pragmatic validity of moral pluralism. Advances in Experimental Social Psychology, 47, 55-130.
Graham, J., Haidt, J., Nosek, B., & Carver, Charles S. (2009). Liberals and Conservatives Rely on Different Sets of Moral Foundations. Journal of Personality and Social Psychology, 96(5), 1029-1046.
Greene, Nystrom, Engell, Darley, & Cohen. (2004). The Neural Bases of
Cognitive Conflict and Control in Moral Judgment. Neuron, 44(2), 389-400. Gruenfeld, D. H., Inesi, M. E., Magee, J. C., and Galinsky, A. D. (2008). Power and
the objectification of social targets. Journal of Personality and Social Psychology, 95, 111–27.
Haidt, J. (2008, March). Jonathan Haidt: The moral roots of liberals and conservatives [Video file]. Retrieved from
https://www.ted.com/talks/jonathan_haidt_on_the_moral_mind
Hall, J. A., Coats, E. J., & LeBeau, L. S. (2005). Nonverbal behavior and the vertical dimension of social relations: A meta-analysis. Psychological Bulletin, 131(6), 898–924.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: Guilford Press.
Keltner, D. (2016). The Power Paradox: How We Gain and Lose Influence. Penguin.
Keltner, D., and Gruenfeld, D. H, and Anderson, C. (2003). Power, approach, and inhibition. Psychological Review, 110, 265–84.
Keltner, D., Van Kleef, G. A., Chen, S., & Kraus, M. W. (2008). A reciprocal influence model of social power: Emerging principles and lines of inquiry. Advances in Experimental Social Psychology, 40, 151–192.
Kipnis, D. (1972). Does power corrupt? Journal of Personality and Social Psychology, 24, 33–41.
Lammers, J., Stapel, D. A., and Galinsky, A. D. (2010). Power increases hypocrisy: moralizing in reasoning, immorality in behavior. Psychological Sci- ence, 21, 737–44.
Lammers, J., Stoker, J. I., Jordan, J., & Pollmann, M. (2011). Power increases infidelity among men and women. Psychological Science, 22, 1191–1197. doi:10.1177/0956797611416252
Lammers, J., Stoker, J., Rink, F., & Galinsky, A. (2016). To Have Control Over or to Be Free From Others? The Desire for Power Reflects a Need for Autonomy.
Personality and Social Psychology Bulletin, 42(4), 498-512
Machiavelli, Niccolo. The Prince: Transl. with an Introd. by George Bull. Penguin Books, 1975.
Magee, J. C. (2009). Seeing power in action: The roles of deliberation,
implementation, and action in inferences of power. Journal of Experimental Social Psychology, 45, 1–14.
Magee, J. C., & Galinsky, A. D. (2008). Social hierarchy: The self-reinforcing nature of power and status. Academy of Management Annals, 2, 351–398.
Stamkou, E., Van Kleef, G.A., Homan, A. C., & Galinsky, A. D. (2016). How norm violations shape social hierarchies: Those who stand on top block norm violators from rising up. Group Processes & Intergroup Relations, 19(5), 608-629.
Stürmer, S., & Snyder, M. (2010). The psychology of prosocial behavior : Group processes, intergroup relations, and helping. Malden, MA: Wiley-Blackwell. Van Kleef, G. A., Homan, A. C., Finkenauer, C., Blaker, N. M., & Heerdink, M. W.
(2012). Prosocial norm violations fuel power affordance. Journal of
Experimental Social Psychology, 48, 937–942.
Van Kleef, G. A., Homan, A. C., Finkenauer, C., Gündemir, S., & Stamkou, E. (2011). Breaking the rules to rise to power: How norm violators gain power in the eyes of others. Social Psychological and Personality Science, 2, 500–507.
Van Kleef, G. A., Oveis, C., van der Lowe, I., LuoKogan, A., Goetz, J., and Keltner, D. (2008). Power, distress, and compassion: turning a blind eye to the suffering of others. Psychological Science, 19, 1315–22.
Appendix A: Supplementary materials
In both studies, we also added for exploratory purposes volition inferences, moral emotions, perceived morality (behavior), warmth, competence and morality (character). An example item for the former is “___ can act the way he wants” (α = .73). For moral emotions, we assessed anger, contempt and disgust (α = .85). For perceived morality in regards to the behavior, participants indicated their agreement (1 = totally disagree, 7 = totally agree) with six adjectives concerning the actor’s behavior. Example adjectives are moral, unethical (reversed) and fair (α = .93). Lastly, the actor’s character was assessed in regards to his warmth (α = .92), competence (α = .89) and morality (α = .91) and participants indicated the extent specific adjectives described the actor. Example items for the three domains are friendly, capable and fair, respectively. A summary of all results, including the supplementary exploratory analysis, can be found in Table 1-4 in the following.
Ta bl e 1: Me an s (a nd s ta nd ar d de vi at io ns ) pe r condi ti on and F -st at ist ic st ud y 1 Co n d it io n F -st at is ti c Va ri ab le Im m or al & Pr os oc ia l Im m or al & Se lf is h Mo ra l & Pr os oc ia l Mo ra l & Se lf is h Ma in e ff ec t (B eh av io r) Ma in e ff ec t (I n te n ti on s) In te ra ct io n Po w er Pe rc ep ti on 1 5. 13 (0. 92) a 4. 83 (0. 75) ab 4. 85 (0. 78) ab 4. 62 (0. 87) b F (200, 1) = 4. 43 , p = .0 37 , η 2 = 0. 02 F (200, 1) = 4. 88 , p = .0 28 , η 2 = 0. 02 F (200, 1) = 0. 10, p = .7 η 2 = 0. 00 Po w er Pe rc ep ti on 2 5. 34 (0. 85) a 4. 64 (0. 91) b 5. 16 (0. 74) ac 4. 87 (0. 81) bc F (200, 1) = 0. 04, p = .8 3, η 2 = 0. 00 F (200, 1) = 17. 92, p < .0 01 , η 2 = 0. 08 F (200, 1) = 2. 99, p = .0 η 2 = 0. 01 Po w er Af fo rd an ce 1 3. 48 (1. 25) a 2. 52 (1. 41) b 4. 86 (1. 12) c 3. 74 (1. 72) a F (200, 1) = 44. 44, p < .0 01 , η 2 = 0. 16 F (200, 1) = 28. 73, p < .0 01 , η 2 = 0. 11 F (200, 1) = 0. 16, p = .6 η 2 = 0. 00 Po w er Af fo rd an ce 2 4. 33 (1. 18) a 3. 12 (1. 38) b 5. 15 (0. 91) c 4. 17 (1. 41) a F (200, 1) = 29. 10, p < .0 01 , η 2 = 0. 11 F (200, 1) = 39. 69, p < .0 01 , η 2 = 0. 15 F (200, 1) = 0. 44, p = .5 η 2 = 0. 00 So ci al En ga ge m en t 4. 07 (0. 99 )a 3. 48 (1. 11 )b 5. 44 (0. 95 )c 4. 21 (1. 08 )a F (2 00 , 1 ) = 5 2. 37 , p < .0 01 , η 2 = 0. 18 F (200, 1) = 39. 54, p < .0 01 , η 2 = 0. 13 F (200, 1) = 4. 70, p = .0 η 2 = 0. 02 Vo li ti on In fe re nc es 4. 44 (1. 18) a 4. 76 (1. 19) a 4. 63 (0. 93) a 5. 00 (1. 11) a F (200, 1) = 1. 86, p = .1 7, η 2 = 0. 01 F (200, 1) = 4. 96, p = .0 27 , η 2 = 0. 02 F (200, 1) = 0. 03, p = .8 η 2 = 0. 00 Mo ra l Em ot io ns 3. 10 (1. 45) a 4. 07 (1. 53) b 1. 88 (1. 09) c 2. 37 (1. 43) c F (200, 1) = 56. 66, p < .0 01 , η 2 = 0. 21 F (200, 1) = 14. 10, p < .0 01 , η 2 = 0. 05 F (200, 1) = 1. 60, p = .2 η 2 = 0. 01 M or al it y (B eh av io r) 3. 63 (1. 05) a 3. 05 (1. 23) b 5. 73 (0. 86) c 4. 89 (1. 31) d F (200, 1) = 156. 95, p < .0 01 , η 2 = 0. 42 F (200, 1) = 20. 42, p < .0 01 , η 2 = 0. 05 F (200, 1) = 0. 61, p = .4 η 2 = 0. 00 Wa rm th 3. 69 (1. 00) a 3. 04 (1. 13) b 5. 21 (0. 85) c 3. 88 (1. 06) a F (200, 1) = 68. 99, p < .0 01 , η 2 = 0. 21 F (200, 1) = 48. 76, p < .0 01 , η 2 = 0. 15 F (200, 1) = 5. 67, p = .0 η 2 = 0. 02 Co m pe te nc e 5. 35 (0. 95) a 4. 64 (1. 11) b 5. 56 (0. 88) a 5. 42 (0. 93) a F (200, 1) = 13. 42, p < .0 01 , η 2 = 0. 06 F (200, 1) = 9. 78, p = .0 02 , η 2 = 0. 04 F (200, 1) = 4. 48, p = .0 η 2 = 0. 02 Mo ra li ty (C ha ra ct er) 4. 06 (1. 15) a 3. 29 (1 .2 5)b 5. 18 (0. 87) c 4. 25 (1. 10) a F (200, 1) = 45. 52, p < .0 01 , η 2 = 0. 17 F (200, 1) = 29. 78, p < .0 01 , η 2 = 0. 11 F (200, 1) = 0. 27, p = .6 η 2 = 0. 00 No te . M ea ns w it h di ff er en t s ub sc ri pt d if fe r at p < .0 5 (T uk ey ’s H on es t S ig ni fi ca nc e D if fe re nc e te st ). Po w er Pe rc ep ti on 1 : G en er al iz ed Se ns e of Po w er s ca le ( A nd er so n, J oh Ke lt ne r, 2 01 2) , P owe r P er ce pt io n 2: r at in g of p owe rf ul a dj ec ti ve s, P owe r Af fo rd an ce 1 : e xt en t b os s, p ol it ic al le ad er & p ro m ot io n an d P ow er A ff or da nc e 2: le ad er sh ip e nd or sc al e (S tam ko u et a l., 2 01 6) .
Ta bl e 2: De sc ri pt iv e st at is ti cs , a lp ha s an d in te r-co rr el at io ns o f t he v ar ia bl es o f s tu dy 1 M SD α 1 2 3 4 5 6 7 8 9 10 11 1 Po w er Pe rc ep ti on 1 4. 90 0. 85 .7 7 2 Po w er Pe rc ep ti on 2 5. 00 0. 87 .6 4 .4 2 ** * 3 Po w er Af fo rd an ce 1 3. 65 1. 62 .9 0 -.0 1 .3 7 ** * 4 Po w er Af fo rd an ce 2 4. 19 1. 42 .8 8 .1 1 .4 2 ** * .8 7 ** * 5 So ci al En ga ge m en t 4. 30 1. 25 .8 4 .0 5 .3 6 ** * .7 1 *** .6 8 ** * 6 Vo li ti on In fe re nc es 4. 71 1. 12 .7 3 .2 3 * .1 5 * -.0 5 -.0 3 -.0 7 7 Mo ra l Em ot io ns 2. 85 1. 61 .8 5 .0 4 -.2 6 ** -.5 3 *** -.5 0 ** * -.5 3 ** * .0 5 8 M or al it y (B eh av io r) 4. 33 1. 53 .9 3 -.0 8 .2 4 ** * .7 1 *** .6 5 ** * .7 0 ** * .0 5 -.6 5 ** * 9 Wa rm th 3. 96 1. 28 .9 2 .0 0 .3 2 ** * .7 0*** .6 9 ** * .8 5 ** * -.1 1 -.5 3 ** * .7 4 ** * 10 Co m pe te nc e 5. 24 1. 03 .8 9 .3 3 ** * .5 6 ** * .4 8 *** .5 3 ** * .5 6 ** * .0 8 -.4 1 ** * .4 4 ** * .5 2 ** * 11 Mo ra li ty (C ha ra ct er) 4. 20 1. 28 .9 1 .0 0 .3 2 ** * .6 7 *** .6 5 ** * .7 7 ** * -.0 6 -.5 6 ** * .7 6 ** * .8 7 ** * .5 9 ** * No te . Po w er Pe rc ep ti on 1 : G en er al iz ed Se ns e of Po w er s ca le ( A nd er so n, J oh n & K el tn er , 2 01 2) , Po w er Pe rc ep ti on 2 : r at in g of p ow er fu l ad jec ti ves , P ow er A ff or dan ce 1: e xt ent bos s, pol it ic al le ade r & pr om ot ion and P ow er A ff or da nc e 2: le ade rs hi p en dor se m en t s ca le ( St am ko u et a l., 2016 ). *** p < .0 01 ** p < .0 1 * p < .0 5
Ta bl e 3: Me an s (a nd s ta nd ar d de vi at io ns ) pe r co nd it io n an d F -st at ist ic st ud y 2 Co n d it io n F -st at is ti c Va ri ab le Im m or al & Pr os oc ia l Im m or al & Se lf is h Mo ra l & Pr os oc ia l Mo ra l & Se lf is h Ma in e ff ec t (B eh av io r) Ma in e ff ec t (C on se q u en ce s) In te ra ct io n Po w er Pe rc ep ti on 1 5. 34 (0. 81) a 5. 09 (0. 61) ab 4. 94 (0. 73) b 5. 06 (0. 68) ab F (2 00 , 1 ) = 4 .5 7, p = .0 34 , η 2 = 0. 02 F (2 00 , 1 ) = 0 .4 4, p = .5 1, η 2 = 0. 00 F (2 00 , 1 ) = 3 .4 0, p = .0 67 η 2 = 0. 02 Po w er Pe rc ep ti on 2 5. 31 (0. 79) a 4. 90 (0. 77) b 5. 29 (0. 67) a 5. 26 (0. 78) ab F (2 00 , 1 ) = 2 .6 9, p = .1 0, η 2 = 0. 01 F (2 00 , 1 ) = 4 .3 0, p = .0 39 , η 2 = 0. 02 F (2 00 , 1 ) = 3 .2 2, p = .0 74 η 2 = 0. 02 Po w er Af fo rd an ce 1 3. 49 (1. 23) a 3. 14 (1. 37) a 4. 87 (0 .9 4)b 4. 55 (1. 17) b F (2 00 , 1 ) = 6 9. 91 , p < .0 01 , η 2 = 0. 26 F (2 00 , 1 ) = 4 .0 1, p = .0 47 , η 2 = 0. 01 F (2 00 , 1 ) = 0 .0 1, p = .9 4, η 2 = 0. 00 Po w er Af fo rd an ce 2 3. 98 (1. 08) a 3. 45 (1. 33) a 5. 22 (0. 79) b 4. 88 (1. 06) b F (2 00 , 1 ) = 7 7. 67 , p < .0 01 , η 2 = 0. 27 F (200, 1) = 8. 01, p = .0 05 , η 2 = 0. 03 F (2 00 , 1 ) = 0 .4 0, p = .5 3, η 2 = 0. 00 So ci al En ga ge m en t 4. 05 (0. 84) a 3. 41 (0. 89) b 5. 22 (0. 94) c 4. 79 (0. 95) c F (2 00 , 1 ) = 1 01 .4 8, p < .0 01 , η 2 = 0. 32 F (2 00 , 1 ) = 1 7. 92 , p < .0 01 , η 2 = 0. 06 F (2 00 , 1 ) = 0 .7 0, p = .4 1, η 2 = 0. 00 Vo li ti on In fe re nc es 4. 70 (1. 14) a 4. 94 (0. 83) ab 5. 08 (0. 95) ab 5. 18 (0. 77) b F (2 00 , 1 ) = 5 .6 7, p = .0 18 , η 2 = 0. 03 F (2 00 , 1 ) = 1 .7 7, p = .1 9, η 2 = 0. 01 F (2 00 , 1 ) = 0 .2 8, p = .6 0, η 2 = 0. 00 Mo ra l Em ot io ns 3. 35 (1. 14) a 3. 93 (1. 20) a 1. 92 (1. 16) b 2. 22 (1. 10) b F (2 00 , 1 ) = 9 5. 07 , p < .0 01 , η 2 = 0. 31 F (2 00 , 1 ) = 7 .4 9, p = .0 07 , η 2 = 0. 02 F (2 00 , 1 ) = 0 .7 6, p = .3 8, η 2 = 0. 00 M or al it y (B eh av io r) 3. 66 (0. 91) a 3. 26 (1. 08) a 5. 63 (0. 96) b 5. 28 (1. 02) b F (2 00 , 1 ) = 2 05 .5 2, p < .0 01 , η 2 = 0. 50 F (200, 1) = 7. 15, p = .0 08 , η 2 = 0. 02 F (2 00 , 1 ) = 0 .0 2, p = .8 9, η 2 = 0. 00 Wa rm th 3. 50 (0. 78) a 2. 90 (0. 96) b 5. 03 (0. 86) c 4. 48 (0. 99) d F (2 00 , 1 ) = 1 50 .4 7, p < .0 01 , η 2 = 0. 41 F (2 00 , 1 ) = 2 0. 94 , p < .0 01 , η 2 = 0. 06 F (2 00 , 1 ) = 0 .0 3, p = .8 7, η 2 = 0. 00 Co m pe te nc e 5. 23 (0. 84) a 4. 71 (1. 13) b 5. 76 (0. 84) c 5. 57 (0. 80) ac F (2 00 , 1 ) = 2 9. 45 , p < .0 01 , η 2 = 0. 12 F (2 00 , 1 ) = 7 .8 2, p = .0 06 , η 2 = 0. 03 F (2 00 , 1 ) = 1 .5 9, p = .2 1, η 2 = 0. 01 Mo ra li ty (C ha ra ct er) 3. 93 (0. 86) a 3. 52 (1. 16) a 5. 22 (0. 83) b 4. 90 (0. 96) b F (200, 1) = 97. 97, p < .0 01 , η 2 = 0. 32 F (2 00 , 1 ) = 7 .3 5, p = .0 07 , η 2 = 0. 02 F (2 00 , 1 ) = 0 .1 5, p = .7 0, η 2 = 0. 00 No te . M ea ns w it h di ff er en t s ub sc ri pt d if fe r at p < .0 5 (T uk ey ’s Ho ne st S ig ni fi ca nc e Di ff er en ce te st ). P owe r P er ce pt io n 1: Ge ne ra li ze d S en se o f P owe r sc al e (A nd er so n, J oh n & K el tn er , 2 01 2) , P ow er P er ce pt io n 2: r at in g of p ow er fu l a dj ec ti ve s, P ow er A ff or da nc e 1: e xt en t b os s, p ol it ic al le ad er & p ro m ot ion an d P ow er A ff or da nc e 2: le ad er shi p en do rs em en t s ca le ( S tam ko u et al ., 2016 ).
T ab le 4: De sc ri pt iv e st at is ti cs , a lp ha s an d in te r-co rr el at io ns o f t he v ar ia bl es o f s tu dy 2 M SD α 1 2 3 4 5 6 7 8 9 10 11 1 Po w er Pe rc ep ti on 1 5. 11 0. 72 .7 9 2 Po w er Pe rc ep ti on 2 5. 19 0. 77 .6 8 .5 6 *** 3 Po w er Af fo rd an ce 1 4. 01 1. 38 .8 9 .0 8 .3 8 *** 4 Po w er Af fo rd an ce 2 4. 38 1. 28 .9 0 .0 7 .4 2 ** * .8 8 *** 5 So ci al En ga ge m en t 4. 37 1. 13 .8 7 .0 7 .3 9 *** .7 0 *** .6 4 *** 6 Vo li ti on In fe re nc es 4. 98 0. 94 .7 7 .3 2 *** .3 6 *** .1 6 * .1 6 * .1 5 * 7 Mo ra l Em ot io ns 2. 85 1. 41 .8 2 .0 7 -.1 8 * -.5 1 *** -.5 1 *** -.5 6 ** * -.1 5 * 8 M or al it y (B eh av io r) 4. 46 1. 42 .9 4 -.0 4 .3 2 *** .7 6 *** .7 3 *** .7 7 ** * .1 4 * -.7 0 ** * 9 Wa rm th 3. 98 1. 22 .9 4 -.0 8 .2 5 *** .6 9 *** .6 7 *** .8 3 ** * .0 4 -.5 4 ** * .7 8 ** * 10 Co m pe te nc e 5. 32 0. 99 .9 2 .3 3 ** * .6 5 *** .6 3 *** .6 2 *** .6 5 ** * .2 1 ** -.4 1 ** * .5 9 ** * .5 9 ** * 11 Mo ra li ty (C ha ra ct er) 4. 39 1. 18 .9 1 .0 3 .4 3 *** .7 9 *** .7 8 *** .7 8 ** * .1 2 -.5 8 ** * .8 4 ** * .8 3 ** * .7 3 ** * No te . Po w er Pe rc ep ti on 1 : G en er al iz ed Se ns e of Po w er s ca le ( A nd er so n, J oh n & K el tn er , 2 01 2) , Po w er Pe rc ep ti on 2 : r at in g of p ow er fu l ad jec ti ves , P ow er A ff or dan ce 1: ex ten t b os s, p ol it ic al lea der & p ro m ot io n an d P ow er A ff or dan ce 2: le ad er sh ip e nd or se m en t s ca le ( S tam ko u et a l., 2016 ). *** p < .0 01 ** p < .0 1 * p < .0 5
