MORAL JUDGMENT AND THE GAINING OR LOSING OF POWER:
WHAT ROLE DOES IDENTIFICATION PLAY IN MORAL HYPOCRISY?
Master thesis, MscBA, specialization Human Resource Management University of Groningen, Faculty of Organizational Behavior
March 9, 2011
LOTTE PIGMANS
Student number: 1475010 A J Lutulistraat 44 9728 WT Groningen tel.: +31 (0)6-50886820 E-mail: l.pigmans@student.rug.nl
Supervisor J. Jordan
ABSTRACT
A possible mechanism behind the phenomenon of moral hypocrisy is identification. In the current study I examined the effects of gaining or losing power on moral hypocrisy and the role of identification in this relationship via a survey study in which I assigned 94 participants to three conditions: expecting to lose power, expecting to gain power, and a control condition.
Participants read vignettes involving high- and low-power transgressors engaging in immoral behavior and rated these transgressions on their morality and their identification with the transgressor and situation. I found that participants in the control group judged immoral behaviors performed by low-power transgressors (marginally) more strictly than the gaining- power group, suggesting feelings of guilt or pity. In addition, I found identification with the transgressor to be positively related to moral judgment for all groups. I discuss the implications of these findings for both theory and organizational practice.
CONTENTS
ABSTRACT ...2
CONTENTS………3
INTRODUCTION ...4
MORAL HYPOCRISY ...5
THEORETICAL BACKGROUND ...8
POWER-GAIN AND MORAL HYPOCRISY ...8
POWER-LOSS AND MORAL HYPOCRISY ...9
IDENTIFICATION ... 10
THE CURRENT RESEARCH ... 13
METHODS ... 14
PARTICIPANTS ... 14
DESIGN... 14
MEASURES ... 14
Moral judgment. ... 14
Identification. ... 15
PROCEDURES ... 15
RESULTS ... 16
MANIPULATION CHECK ... 16
MORAL JUDGMENT ... 17
IDENTIFICATION ... 18
MEDIATION BY IDENTIFICATION ... 18
DISCUSSION... 20
REFLECTION OF RESULTS ... 20
Moral judgment. ... 20
Identification. ... 20
Mediation by identification. ... 21
LIMITATIONS OF STUDY ... 21
FUTURE RESEARCH ... 22
PRACTICAL IMPLICATIONS... 23
CONCLUSIONS ... 23
REFERENCES ... 24
INTRODUCTION
The fraud of Enron, the adultery of Bill Clinton and Tiger Woods, as well as mundane transgressions like lying, stealing, and speeding in traffic, are all examples of immoral behavior on which individuals have an opinion. While it might seem that everyone condemns immoral behavior, this assumption is not entirely true. Individuals’ judgments of others’ moral behavior are dependent on several variables (e.g., Hauser, 2006; Klinger, Albaum & Hetherington, 1964;
Mikhail, 2007; Pizarro, 2000). First, there is the behavior that is the target of judgment (Mikhail, 2007; Rest, 1986). For example, few individuals morally condone a man physically abusing his wife. But many people likely condone a parent verbally disciplining his child.
Second, the context in which the behavior occurs influences the way individuals judge it (Kohlberg, 1984). If the mentioned man’s wife committed adultery, this situational attribute may lead to a more lenient judgment of the man’s behavior by some individuals (but not by others) – which brings us to the third variable: the effect of the individual perceiver (Forgas, 1988; Narvaez & Hill, 2010; Trevino, 1986). The effect of the act and the context it is in may be different for individuals with different life experiences and perspectives. In the example above, someone who has been cheated on himself may have an entirely different judgment of adultery from someone who works in a shelter for abused women. Lastly, there is also the effect of the transgressor and his or her relationship with the person who is exercising judgment. For example, an immoral behavior committed by someone in a person’s in-group (e.g., a close friend) is likely to be judged more leniently than is a behavior committed by someone in a person’s out-group (e.g., a distant stranger) (e.g., Brewer, 1979; Moghaddam &
Stringer, 1988; Mullen, Brown & Smith, 1992).
These examples are fictitious; however, previous research supports the proposal that, like in the examples, the judge’s mindset and the (relationship with the) person being judged play an important role in how immoral behaviors are perceived (e.g. Forgas, 1998, Greene, Sommerville, Nystrom, Darley & Cohen, 2001; Greene & Haidt, 2002; Loewenstein, Weber, Hsee & Welch, 2001; Pizarro, 2000). Therefore, there are significant differences in the leniency with which individuals judge others and themselves when it comes to immoral behavior.
The fact that mindset (or one could also consider it, perspective) of the perceiver and characteristics of the transgressor play an important role in moral judgment, suggests that our judgment is influenced by factors that cannot strictly be assigned to the act that is being judged.
However, while previous research has examined the role of mindset of the perceiver (Lammers
et al., 2009) or characteristics of those being perceived (e.g., Haidt & Graham, 2007; Reed &
Aquino, 2003; Valdosolo & DeSteno, 2007) on moral judgment, no prior studies have looked at these two factors in concert, nor on how they interact. The current study aims to do this.
Specifically, the perspective-related factor I am studying in this research is the perceiver’s own closeness to gaining (which, I will label close to power-gain) or losing power (which, I will label close to power-loss). And the target-characteristic that I am studying in this research is the power-level of the transgressor. I propose that these two factors will interact to affect individuals’ moral judgment of the acts in question. I explain my hypotheses and the rationale behind these hypotheses in detail in the sections that follow.
Moral Hypocrisy
Within the domain of moral judgment, this study is focusing in particular on the topic of moral hypocrisy. Moral hypocrisy has been conceptualized as an individual’s ability to hold a belief while acting in discord with it (Batson, Kobrynowicz, Dinnerstein, Kampf, & Wilson, 1997). According to this definition, individuals hold certain moral standards and say that they hold themselves and others towards these standards - but do not act in accordance with the standards. However, there also is another definition of moral hypocrisy. In this definition, individuals actually hold different moral guidelines for the self than they do for others (Valdosolo & DeSteno, 2007), meaning that it they do not profess having egalitarian standards across individuals but rather they explicitly state that the self is subject to more liberal moral standards than are others. Thus, one could consider this a more blatant manifestation of moral hypocrisy. As Valdesolo and DeSteno (2007) found, this bias extends to group-level social identities, meaning that individuals apply lower moral standards to those in their in-group but not to those in their out-group.
I assert that losing and gaining power will differentially affect moral judgment towards those low or high in power based on the differential identification with the transgressor that such roles produce. Identification with the transgressor has been shown to influence moral judgment (Valdesolo & DeSteno, 2007). For example, several studies have shown that individuals make more lenient judgments others of the same sex (Klinger et al., 1964; Maccoby
& Wilson, 1957). Social status is another dimension on which individuals identify with others.
For example, Maccoby & Wilson (1957) found that individuals identified more closely with movie stars who belonged to the social class they aspired to belong to.
Despite the large body of literature that has been published on power in the last decade (e.g., Greer & Van Kleef, 2010; Handgraaf, Van Dijk, Vermunt, Wilke & De Dreu, 2008;
Raven, 2008), none of this research has examined power-level as a source of identification.
However, the possibility has not gone unnoticed. For example, Lammers, Stapel, and Galinsky (2010) suggested that power might be a viable source of identification. They found that individuals judged the morality of others at the same power-level as themselves differently from those who were at a different power-level. They labeled these inconsistent judgments, moral hypocrisy and moral hypercrisy. I will explain both of these terms in detail later in this thesis.
Lammers and colleagues (2010) explained the phenomenon of moral hypocrisy as being the result of feelings of entitlement of high-power individuals, and of a focus on rewards and a reduced sensitivity to social disapproval. Although they did not explicitly name identification based on power role as the mechanism for such differences in judgment, identification is a viable explanation. In support of this reasoning, Valdesolo and DeSteno (2007) found an effect of ‘in-group morality’ on moral judgment. This phenomenon was such that individuals judged more leniently on themselves and others who they perceived as in-group members. And these findings appear to be consistent with other research that has shown that individuals judge more positive and leniently on others they perceive as similar (Fensterheim & Tresselt, 1953; Klinger et al., 1964).
Whereas Lammers and colleagues (2010) examined individuals who were already in a certain power position, I will examine the influence of expecting to obtain a certain power (or powerless) position in the future on people’s moral judgments of others. The question is how individuals judge others that belong to a group that they were coming from versus those within the group that they will be a part of in the near future and whether this relationship changes based on whether the past or future group is of low- or high-power. I will also examine if and how group identification affects these effects. Thus, the primary research questions in this study are:
How does the expectation of power-gain and power-loss affect moral judgment of others in a high- or low-power position?
What role does identification play in this relationship?
In the following section I review the relevant literature on power, moral judgment, and identification. I also elaborate on these concepts in greater detail and present the resulting hypotheses. Third, I explain how I tested these hypotheses in a survey study amongst students
and recent graduates. Fourth, I present the results of this study and discuss them in the context of the current hypotheses. Lastly, I discuss implications for practice and theory.
THEORETICAL BACKGROUND
Power-gain and Moral Hypocrisy
Different definitions of power can be found throughout the literature. A distinction can be made between power over one’s own actions (i.e., personal power) and power over others (i.e., social power) (Lammers, Stoker & Stapel, 2009). In line with most previous studies (Dahl, 1957; Keltner, Grunfeld & Anderson, 2003; Salancik & Pfeffer, 1974; Wrong, 1968), in the current research, I operationalize power as, power over others. Thus, the definition of power used in this study focuses on ‘an individual’s intentional and effective capacity to control, modify or influence others’ (Galinsky et al., 2003, p. 3).
Power is attractive: most individuals prefer a position with more power to a position with less power (Mulder, 1977). Individuals who have less power try to acquire more power and individuals who already have power try to protect and augment it (Sivanathan, Pillutla &
Murnighan, 2007). Sivanathan and colleagues (2007) showed that individuals tend to overreact to the gaining of power by increasing their demands to a level that is greater than reasonable for the power-level they rise to. Reactions on the expectation of gaining power then seem to be determined by the desire for power. This makes it likely that the same effect of power on moral judgment that Lammers and colleagues (2010) found for individuals in a high-power position will also be found for close to power-gain individuals. In other words, I expect that as opposed to close to power-loss individuals (as well as those who do not expect to gain or lose power), close to power-gain individuals will display moral hypocrisy.
Lammers and colleagues (2010) tested for the effect of moral hypocrisy using Batson et al.’s (1997) definition of acting in discord with one’s own beliefs, as well as Valdesolo and DeSteno’s (2007) definition of holding different standards for the self versus others. They found that individuals in power practiced both forms of hypocrisy. For example, in line with Batson and colleagues’ (1997) definition, they compared the morality of participants’ own transgressions in an experiment to their judgment of others’ transgressions. Also, in line with Valdesolo and DeSteno’s definition, they asked participants in different experiments to rate their own and others’ transgressions. They found that high-power individuals not only showed behavior that differed from their own moral standards, but also that they held different standards for themselves as for others. Although Lammers and colleagues did not test for individuals that were close to gaining power, the increase in demands by individuals who gained power (Sivanathan et al., 2007) suggests that the presence of a moral hypocrisy in close
to power-gain individuals may be even more pronounced than that found in individuals already holding a high-power position.
Based on these previous findings, I hypothesize that close to power-gain individuals, as opposed to a control condition, will judge those in a high-power position more leniently. In contrast, based on the theory of moral hypocrisy (Lammers et al., 2010; Valdesolo & DeSteno, 2007), close to power-gain individuals are expected to judge no more leniently or harshly on transgressors in a low-power position compared to a control condition. Thus, they will not show more lenient judgment towards others who are not members of their perceived in-group.
Hypothesis 1: Those close to power-gain will judge those in a high-power position more leniently than will those in a control group.
Hypothesis 2: Those close to power-gain and a control group will judge those in a low- power position equally.
Power-loss and Moral Hypocrisy
Whereas the prospect of gaining power seems attractive, this reality clearly is not true for the prospect of losing power. Sivanathan and colleagues (2007) found that relative to those individual who experienced a sudden gain in power, individuals who lost power lowered their demands to an appropriate level. Thus, unlike the power-gainers, they did not overcompensate for their change in status. This appropriate reaction seems to suggest that identification with individuals in a high-power position will diminish or disappear for close to power-loss individuals, since they are no longer part of the powerful in-group.
When an individual has the expectancy of losing power, he or she might experience a feeling of rejection by the high-power group. Theory and research show that group-rejection leads to a decrease in liking of members of the in-group (Nesdale, Durkin, Maass, Kiesner, Griffiths, Daly & McKenzie, 2010). Thus, it seems likely that group-rejection will lead to a more severe moral judgment of members of one’s former in-group.
Another theory that seems to apply to the situation for close to power-loss individuals is cognitive dissonance (Festinger, 1957). This theory states that if a person holds two cognitions that are inconsistent with one another, he or she will seek to remove the pressure of cognitive dissonance, among other ways, by removing one of the two ‘dissonant’ cognitions. Adding new cognitions that are consonant with the knowledge or behavior an individual hopes to maintain is one way to reduce dissonance. Because they are fungible, cognitions that the most often change are the ones that are connected to attitudes, beliefs and opinions (Kretchmar,
2008). If a person is expecting to lose power, this expectation might lead to a change in attitude towards individuals in a high-power position. It seems likely that an individual in such a position would have a more negative attitude towards individuals in a high-power position, as a way of psychologically coping with the impending power-loss. This change in attitude might lead to a more severe moral judgment, and a decrease in, or disappearance of identification with this group. Thus, close to power-loss individuals are expected to judge the moral actions of high-power targets more harshly than are individuals who are not facing an impending loss of power.
Hypothesis 3: Those close to power-loss will judge those in a high-power position more harshly than those in a control group.
Whereas the expectancy of power-loss seems to lead to a more severe moral judgment for individuals in a high-power position, relative to the judgments of those in a control condition, as stated above, I do not expect moral hypocrisy to be manifest for the judgment of low-power individuals’ moral transgressions.
Lammers and colleagues (2010) found an effect of moral hypercrisy for low-power individuals, that is, they were more severe in the judgment of their own moral transgressions. If close to power-loss individuals consider themselves as being in a low-power position, this effect may be expected to appear for them, as well. However, as I will discuss later, I predict that close to power-loss individuals do not identify with others in a low-power position. Thus, I do not expect the expectancy of power-loss to influence identification and thereby moral hypocrisy in moral judgment of individuals in a low-power position.
Hypothesis 4: Those close to power-loss and a control group will judge those in a low- power position equally.
Identification
As I stated above, for close to power-gain individuals, I predict that their desire to gain power, that is to be part of the high-power group, will lead them to engage in moral hypocrisy.
For close to power-loss individuals, I expect that group-rejection and cognitive dissonance (Festinger, 1957; Kretchmar, 2008; Nesdale et al. 2010) will lead to a decrease or disappearance in moral hypocrisy – towards either the transgressions of both low- or high- power targets, even leading to more harsh judgments of those in high-power positions.
However, neither of these predictions directly address the mechanism question, namely, what makes moral hypocrisy occur (or not occur)? Consistent with the finding that individuals judge others they perceive as in-group members more leniently than they do those perceived as out- group members (Valdesolo & DeSteno, 2007), I propose that this mechanism is power-role identification.
A good deal of research on identification exists in the social psychology literature.
Theory and research exists on both why individuals identify with certain others and on what effects this identification has. In both research and theory on identification, perceived similarity on various aspects of the other is found to be the basis for identification processes to occur (e.g.
Riordan, Schaffer & Steward, 2005; Tajfel & Turner, 1986; Tsui, Egan & O’Reilly, 1992).
The most important theory on identification is the social identity theory (Tajfel &
Turner, 1979, 1986). According to this theory, individuals define themselves in terms of social and demographic categories such as ethnicity or gender. Extending this work to the current study, power group could indeed be a category with which individuals identify. Research on power proposes that power is a mindset that transforms the way individuals act, think about themselves, and think about others (e.g., Keltner et al., 2003). The central idea of the social identity theory is that individuals seek to obtain a positive evaluation of oneself and part of this self-evaluation depends on belonging to social groups and the evaluations associated with them (Tajfel & Turner, 1979, 1986, as cited in Goldberg, Riordan, & Schaffer, 2010).
There are two dimensions within social identity theory. The first is a cognitive dimension of social classification: human beings tend to simplify the environment into groups of people according to the perceived similarities related to their actions, intentions, or behavior.
One could conceivably perceive a sense of belongingness to others who share his or her power role and view others in such a power position as in-group members. The second dimension within social identity theory is an affective dimension: this dimension corresponds to the need of individuals to have a positive image of themselves derived from positive evaluations linked to belonging to a group. This need for positive self-evaluation suggests that individuals will tend to judge (themselves and) other in-group members more positively. It also suggests that negative (i.e., immoral) actions of other in-group members will be judged more leniently.
In line with the finding that individuals identify with others that belong to a social status group they aspire to belong to themselves (Maccoby & Wilson, 1957), I expect close to power-gain individuals to identify with others in a high-power position because they aspire to be in this group. However, for close to power-loss individuals, I expect the effects of group- rejection and cognitive dissonance (Festinger, 1957; Kretchmar, 2008; Nesdale et al., 2010) to
lead to a decrease in identification with the powerful, and perhaps, as suggested by the group- rejection literature, a complete rejection of individuals in this group. In line with theory on in- group morality, that is the using of different moral standards for individuals considered as in- group members (Haidt & Graham, 2007), I predict that these changes in identification will influence the manifestation of moral hypocrisy for both groups. Thus, I expect a decrease in identification with a group to lead to a decrease in moral hypocrisy (and even to more harsh judgments) and an increase in identification to lead to an increase in moral hypocrisy. As I explained, for individuals in a high-power position, I expect an increase in identification for close to power-gain individuals and a decrease in identification for close to power-loss individuals. Therefore, I hypothesize:
Hypothesis 5a: Close power-gain individuals will identify more than close to power-loss individuals with individuals in a high-power position
Although identification with individuals in a high-power position is expected to diminish for close to power-loss individuals, it is uncertain whether identification with others in a low-power position will arise. Past research shows that individuals identify with groups they aspire to belong to (Maccoby & Wilson, 1957) and it does not seem likely that for close to power-loss individuals, being part of a group of others who are low in power is considered attractive. Thus, I hypothesize:
Hypothesis 5b: Close power-gain individuals and close to power-loss individuals do not differ in identification with others in a low-power position - showing low identification across groups.
In line with theory on in-group morality, that is, the use of different moral standards for individuals considered as in-group members (Haidt & Graham, 2007), I predict that these changes in identification will influence the manifestation of moral hypocrisy for both groups.
Thus, I expect a decrease in identification with a group to lead to a decrease in moral hypocrisy and an increase in identification to lead to an increase in moral hypocrisy. As I explained, for individuals in a high-power position, I expect an increase in identification for close to power- gain individuals and a decrease in identification for close to power-loss individuals. Therefore, I hypothesize:
Hypothesis 6: Close to power-gain individuals’ more lenient judgment of high-power individuals will be mediated by their (greater) identification with these individuals.
Hypothesis 7: Close to power-loss individuals’ more harsh judgment of high-power individuals will be mediated by their (lesser) identification with these individuals.
The Current Research
Based on identification theory (e.g. Goldberg et al., 2010; Maccoby & Wilson, 1957;
Riordan et al., 2005; Tajfel & Turner, 1986; Tsui et al., 1992; Valdesolo & DeSteno, 2007), a new theory can be formed on how being close to gaining or losing power will affect moral judgment of others in high- or low-power positions via identification with the transgressor.
As explained above, theory on power-gain suggests that individuals expecting to gain power will identify with others in a high-power position. Although they are not yet part of the group, I expect that close to power-gain individuals will identify with individuals at a high- power position and feel like they belong to this group rather than to a low-power group (the group they are coming from). For close to power-loss individuals, I predict that these people will reject their former high-power group members, thereby experiencing reduced identification with those high in power and judging them more harshly, but that they will also not show increased identification with those in low-power. Taken together, I predict that these groups will engage in increased and decreased moral hypocrisy towards those high in power, respectively. I also predict that such differences in moral judgment will not occur towards low- power targets because neither close to power-gain individuals nor close to power-loss individuals are expected to identify with this group. I examine these hypotheses in the current study.
METHODS
Participants
I distributed a survey to 94 students and recent graduates of the Hanze University and at the University of Groningen. In exchange for participation, they received entry into a raffle for a 50 Euro IKEA gift card. I instructed them that they would be participating in two studies: one about people’s experiences and their reactions to them, and one in which they would have to give their opinion on behavior of people in different contexts. To recruit participants, I posted a link on the online guestbook of two sports clubs. Also, I approached students at the canteen of the faculty of social sciences. This last set of participants completed the survey via a paper-and- pencil format.
Of the 94 respondents, four had to be removed because they did not complete the experimental prime. In addition, five participants had to be removed because they did not complete all the questions. This resulted in 85 respondents on which to perform the analyses.
Of the participants, 73% were female. They were, on average, 23-years-old (SD = 2.6).
Design
To examine whether the expectation of power-loss or power-gain influenced the moral judgments of those in high or low positions, I used three experimental conditions: being close to gaining power (close to power-gain), being close to losing power (close to power-loss), and a control condition. I had concerns that the mere prospect of a change in power-level would influence judgment of others. Therefore, the control condition was about simply being with others, and thereby neutral when it came the constructs of power and changes in power.
Measures
Moral judgment. To measure the construct of moral judgment, I asked respondents to read and rate the morality of the behavior of persons engaging in immoral behavior on a 7- point Likert scale (1 = not at all; 7 = very much). Two of the stories were about a person high in power, meaning a person with intentional and effective capacity to control, modify or influence others, and two of the stories were about a person low in power, meaning a person without intentional and effective capacity to control, modify or influence others (Dahl, 1957;
Keltner et al., 2003; Salancik & Pfeffer, 1974; Wrong, 1968). Based on Lammers and colleagues (2010) measure of moral judgment, I asked: ‘To what extent do you find this behavior acceptable?,’ ‘To what extent do you find this behavior justifiable?,’ ‘To what extent
do you think this person should be forgiven for his or her behavior?,’ and ‘To what extent do you feel that this person should be punished for his or her behavior?’ For this last question, I reverse-scored the response option values to fit the overall scale. These items formed scales both for the transgressors in a high-power position and the transgressors in a low-power position. I performed a reliability analysis for these two scales. The reliability for the high- power transgressor items was high, α = 0.81. The mean score on this scale was 2.23 (SD = 0.64). The reliability for the low-power transgressor items was adequate, α = 0.66. The mean score on this scale was 2.27 (SD = 0.74).
Identification. I measured the construct of identification using two items. I asked the respondents to rate on a 7-point Likert scale (1 = not at all; 7 = very much) how much they identified (1) with the transgressor and (2) with the situation this person was in. I made an identification scale for the transgressors in all four scenarios, that is (as was also done with the moral judgment items), there was a scale corresponding to each of the four targets. I averaged across the two high-power target scales and the two low-power target scales. I then performed a reliability analysis for each of these two scales. The reliability for the high-power transgressor items was acceptable, α = 0.69. The mean score on this scale was 2.27 (SD = 0.74). The reliability for the low-power transgressor items was comparable, α = 0.63. The mean score on this scale was 2.72 (SD = 1.10). Thus, on average, participants showed a fairly low (that is, below the mid-point of the scale) identification with both high- and low-power transgressors.
Procedures
First, I primed study participants with one of three conditions: two experimental conditions and one control condition. In the first condition (close to power-gain), I asked participants to recall and write about a time when they were expecting to gain power. In the second condition (close to power-loss), I asked participants to recall and write about a time when they were expecting to lose power. A third group of participants was in the control condition. I asked this group to recall a time that they were with other people. I divided participants randomly over the three groups. There were 30 participants in each group.
After priming, I asked all participants to rate four vignettes about persons engaging in immoral behavior (two high-power and two low-power). I then asked participants a variety of questions about this behavior and about individuals’ reactions to the wrong-doing. Also, I asked participants to what degree they identified with both the transgressor and the situation described.
RESULTS
The correlations between all study variables are displayed in Table 2.
Table 2: Correlation matrix of all variables used in analyses
Variables Gender Age Identification HP Judgment HP Identification LP Judgment LP
Gender -
Age 0,00 -
Identification HP 0.22* -0.04 -
Judgment HP -0.07 -0.09 0.40* -
Identification LP -0.01 -0.20 0.46* 0.16* -
Judgment LP 0.01 -0.27* 0.33* 0.41* 0.49* -
Notes: *p ≤ .05; N = 85; HP = high-power; LP = low-power
I checked for gender effects on the main dependent variables of moral judgment and identification, but found no significant effects, all F’s < 1.91, p’s > .16.
Manipulation Check
To check if my manipulation worked in the intended way, I asked participants to indicate the extent to which they expected to gain or lose power in the situation they described earlier (1= lose a lot of power; 7 = gain a lot of power). The experimental conditions affected participants’ feelings of losing or gaining power, F(2, 82) = 74.61; p < .01. As expected, participants in the close to power-gain condition, showed the highest expectation of gaining power, whereas the participants in the close to power-loss condition, showed the lowest expectation of gaining power. However, not all participants experienced the feelings that I expected: seven participants in the close to power-gain condition and three participants in the close to power-loss condition stated that they did not expect to gain or lose power in the situation they described in the manipulation (i.e., they provided a score of 4 or lower if in the close to power-gain condition and of 4 or higher if in the close to power-loss). Thus, I deleted them from the analyses. For the control group, I used only those participants who reported to have not felt to gain or lose power or to gain or lose just a little power (i.e., gave a rating between 3 and 5 on the 7-point scale). I deleted three participants in the control condition who did not meet these criteria. After deleting participants with missing values, this left me with 75
participants on which to perform the analyses. Means and standard deviations for the three conditions are displayed in Table 3.
Table 3: Mean and standard deviation of manipulation check
Condition Mean (SD)
Gaining power 5.04 (0.88)a Losing power 2.41 (0.78)b Control 4.04 (0.79)c
Notes: N = 85. Means with different subscripts are significantly different from each other at p ≤ .05. These means are based on all participants prior to exclusions.
Moral Judgment
There was no difference between groups in their judgment of the high-power transgressors’ behavior, F(2, 71) = 0.43; p = .66. Thus, Hypothesis 1, which predicted that close to power-gain individuals would be more lenient in their moral judgment of high- powered targets relative to those in the control condition, and Hypothesis 3, which predicted that those close to power-loss would judge high-power targets more harshly than those in a control condition, must be rejected.
There was an unexpected (marginal) difference between groups in moral judgment of the low-power transgressors, F(2, 71) = 2.62; p = .08. Post-hoc contrasts showed that, contrary to Hypothesis 2, participants in the close to power-gain condition and control condition differed significantly in their judgment of low-power targets, t(71) = -2.21; p = .03. However, confirming Hypothesis 4, close to power-loss condition and control condition participants indicated an equal moral judgment of those in the low-power condition, t(71) = -1.61; p = .11.
Interestingly, participants in the control condition indicated a lower moral judgment of the low- power target than did participants in both the close to power-gain and close to power-loss conditions taken together, t(71) = -2.22, p = .03. I discuss the potential reasons for this difference, as well as its implications in the Discussion Section. Means and standard deviations for the three conditions are displayed in Table 4.
Table 4: Mean and standard deviation of moral judgment of low and high-power transgressors
Condition High-power Low-power
Gaining power 2.14 (0.61)a 2.74 (0.55)b
Losing power 2.21 (0.66)a 2.63 (0.52)b,c
Control 2.31 (0.67)a 2.57 (0.59)c
Notes: N = 74. Means with different subscripts within each column are significantly different from each other at p ≤ .05
Identification
Hypothesis 5a predicted that close to power-gain individuals would identify more than close to power-loss individuals with individuals in a high-power position. However, there was no difference between groups in their identification with the high-power transgressor: F(2, 75)
= 0.31, p = .74, rejecting Hypothesis 5a. Supporting Hypothesis 5b, which predicted that close to power-gain individuals and close to power-loss individuals would not differ in their identification with those in a low-power position, there was no difference between groups in identification with the low-power transgressors: F(2, 75) = 0.49; p = .62. Means and standard deviations for the three conditions are displayed in Table 5.
Table 5: Mean and standard deviation of identification with high- and low- power transgressors
Condition High-power Low-power
Gaining power 2.26 (0.73) 2.66 (1.59)
Losing power 2.20 (0.71) 2.87 (1.05)
Control 2.36 (0.79) 2.59 (1.09)
Note: N = 74
Mediation by Identification
Hypotheses 6 and 7 predicted that close to power-gain participants’ more lenient judgment of high-power transgressors and close to power-loss participants’ more harsh judgment of high-power transgressors would be explained by these groups’ greater and lesser identification with high-power transgressors, respectively. However, with the non-significant results in the ANOVA analysis for judgment of transgressions committed by high-power individuals, my data did not meet the first criterion for mediation, rejecting Hypothesis 6 and 7:
there should be a significant relationship between the independent (i.e., close to power-gain or
-loss) and the dependent (i.e., moral judgment) variable. (Baron & Kenny, 1986). Therefore, I did not test identification as a potential mediator in either the close to power-gain’s more lenient judgment of the powerful, nor the close to power-loss’ more harsh judgment of the powerful.
However, as an exploratory analysis, I did examine the relationship between identification with the transgressor and moral judgment, absent of the close to power-gain or – loss role of the perceiver. Significant relationships between identification and moral judgment existed for both high-power transgressors, = 0.36, t(73) = 3.23, p < .01, and low-power transgressors, = 0.50, t(73) = 4.89, p < .01. These analyses demonstrated that individuals who expressed greater identification with the transgressor were then more likely to express more positive judgments of his or her behavior. I discuss this finding in more detail in the Discussion.
DISCUSSION
Reflection of Results
Moral judgment. Contrary to my expectation, neither close to power-gain individuals nor close to power-loss individuals differed from the control group in moral judgment of those in high-power positions. As I expected, no differences were found between close to power-loss individuals and the control group in their judgments of those in a high-powered position. A possible explanation for these findings lies in the underlying effect of identification, as I will discuss below.
Unexpectedly, I found a (marginally) lower moral judgment of the low-power target by participants in the control group than by close to power-gain individuals. Since no effects were found between the power groups for identification, an explanation for this finding must be searched elsewhere. I did not find a theory that explains why a more lenient judgment for low- power individuals would occur for individuals that expect to gain power. However, it is plausible that someone who expects to gain power might feel guilty for her good future prospects with regards to those that are in a low-power position (Wayment, 2004), as well as feel pity towards these people. These feelings of guilt or pity then may explain these results.
Identification. The results of this study show that, contrary to the expectations, there is no effect of being close to gaining or losing power on identification with individuals in a high- power position. I found no differences between the two groups in identification with the high- powered targets. A possible explanation may be that close to power-loss individuals still identify more with individuals in a high-power position than I expected. Support for this theory can be found in theory on control (Langer, 1975). As the definition of power shows, control is the essence of power. Literature on control loss shows individuals are reluctant to admit they have lost control (Rodin & Langer, 1980) and react negatively to losses in control. Findings on losing control show that the illusion of control, that is the unjust thinking that and acting like one still is in control, is particularly strong when the sense of control is undermined (Thompson & Armstrong, 1998). Although findings of Sivanathan and colleagues (2007) do not support my proposal that theory on control loss applies to power-loss, they argue themselves that these finding might result from the fact that power-loss in their study is unambiguous. In the current study, it seems more possible to live in denial of the losing of power.
For transgressors in a low-power position, I found no differences between close to power-loss individuals and close to power-gain individuals in identification. These results are
in line with my hypothesis that neither close to power-gain individuals, nor close to power-loss individuals identify with others in a low-power position.
Mediation by identification. Although, due to the non-significant effect between power-gain and –loss and identification, identification could not be examined as a mediator, I did find a relationship between identification and moral judgment, such that greater identification with the target predicted more lenient moral judgment. These findings seem to suggest that high identification leads to more lenient moral judgment. However, previous research has found that similarity of attitudes influences attraction to others (Byrne, 1969): that is, individuals like others with values similar to their own. This suggests that moral judgment will be influenced by identification with the transgressor. In addition, it is also plausible that the found relationship may reflect an effect of moral judgment on identification: correlation does not imply directionality. It is possible that one’s positive moral judgment of another makes him or her identify more strongly with the target of the judgment. Suggestions for future research on this subject are discussed below.
Limitations of this Study
As with any research, this study had its limitations. First, I derived the results through a survey study. This survey began by asking participants to recall a situation in which they expected to gain or lose power. Such a manipulation may lack realism. In addition, the comments on the study showed that participants found it difficult to think of a moment in which they were close to losing power, gaining power, or were just with other people. A lot of people started the survey, but stopped at this question. This is another indication that the question was difficult for participants. However, the manipulation check confirmed that most participants were able to recall the intended situation. This point considered, a manipulation in the lab, in which participants are actually made to be close to gaining or losing power, might have stronger effects than the current recall task.
Another concern is the direction of the positive relationship found in the present study between identification and moral judgment. As mentioned, previous research found an effect of moral judgment on identification (Byrne, 1969). In the present study, I asked questions on identification after participants read the story on the transgressors and answered some questions on moral judgment. Identification, then, may be affected by individuals’ moral judgment of the transgressor’s behavior, instead of the other way around. A goal of this study was to see whether individuals judge more leniently on others they identify with based on power-level. It is possible that this relationship is not what the results reflect. In retrospect, I
should have taken into account the possible effects of moral judgment on identification by possibly counter-balancing the ordering of these questions so that some participants answered the moral judgment questions first and others answered the identification questions first.
Future Research
From the results of this study, some suggestions for future research arise. First, the unexpected finding of a small difference between the control group and the close to power- gain group on the judgment of low-powered individuals needs further study to confirm results.
As discussed above, perhaps individuals that expect to gain power feel guilty about this gain in power and therefore judge those low in power more leniently in order to relieve this guilt.
Another possibility is that close to power-gain individuals pity others in a low-power position.
More research on the mechanism behind the relationship between expecting to gain power and moral judgment of low-power individuals is required.
The second finding that may be studied more thoroughly is that of the relationship between identification and moral judgment. Although previous studies suggest that identification will lead to a more lenient moral judgment (Valdesolo & DeSteno, 2007; Haidt
& Graham, 2007), other studies suggest an effect of moral judgment on identification (Byrne, 1969). In future research it is important to determine the direction of the relationship.
Therefore, questions on identification with individuals that are low or high in power may be presented before describing the moral dilemma. This way, it is ensured that identification is based on power level, and not on one’s moral judgment of the behavior, nor on the behavior itself.
A third suggestion for future research lies in the way the data was examined. In this study, I compared individuals close to power-gain and power-loss on identification and moral judgment, because I expected differences between those groups in identification with high- power transgressors, but not with low-power transgressors. To see whether identification based on power-level was the mechanism behind moral hypocrisy, I wanted to study the effects of differences in identification on moral judgment of a group. However, a possibility lies in using a mixed design of between- and within-subject analyses, to see whether expecting to lose or gain power leads to a different judgment of low-power individuals than of high-power individuals. A close to power-gain individual may judge a low-power individual more harshly than a high-power individual in moral dilemmas, where the reversed may be true for a close to power-loss individual. To see whether these relationships exist, further research is needed.
Practical Implications
The findings of this study lead to some practical implications for organizations. First, if close to power-gain individuals judge more leniently on others in a low-power position due to feelings of guilt or pity, this judgment may affect those close to power-gain’s performance and moral behavior within an organization. Take the following example: a manager is told to she will get a promotion within a short period. This promise may cause that this individual to now hold lower moral standards for her subordinates than was the case in the past. Consequently, subordinates of the mentioned manager may now get away with behavior that the organization does not find acceptable. Therefore, in promotions or other ways of extending power of employees, organizations should manage the moral standards of these employees.
Second, the finding that moral judgment and identification are related, may affect performance within an organization. If individuals judge the moral behavior of those they identify with more leniently, this relaxed judgment may influence their behavior towards each other. For example, it may influence performance evaluations of employees, which lead to the risk of an evaluation not reflecting actual performance. Organizations should keep this in mind in interpreting results of subjective evaluations.
Conclusions
Returning to the research question, the expectation of gaining power does affect moral judgment. However, identification does not seem to play a role in this relationship. Where I expected the expectation of gaining or losing power to lead to moral hypocrisy as a result of this identification, results of this study suggest a different effect. The more lenient moral judgment by close to power-gain individuals of low-power individuals, suggests an effect of feelings of pity or guilt.
REFERENCES
Baron, R.M. & Kenny D.A. 1986. The moderator-mediator variable distinction in social- psychological research: conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51(6): 1173-1182
Batson, C.D., Kobrynowicz, D., Dinnerstein, J.L., Kampf, H.C., & Wilson, A.D. 1997. In a very different voice: unmasking moral hypocrisy. Journal of Personality and Social Psychology, 72: 1335–1348.
Brewer, M.B., 1979. In-group bias in the minimal intergroup situation: a cognitive- motivational analysis. Psychological Bulletin, 86(2): 307-324
Byrne, D. 1969. Attitudes and attraction. In Berkowitzl, L. (Ed.), Advances in Experimental Social Psychology, 4: 35-89, New York: Academic Press
Dahl, R. 1957. The concept of power. Behavioral Science, 2: 201–215
Festinger, L. 1957; renewed in 1985. A theory of cognitive dissonance. Stanford university press
Forgas, J.P. 1998. On being happy and mistaken: mood effects on the fundamental attribution error. Journal of Personality and Social Psychology, 75: 318-331
Galinsky, A. D., Gruenfeld, D. H, & Magee, J. C. 2003. From power to action. Journal of Personality and Social Psychology, 85: 453-466.
Goldberg, C.B., Riordan, C. & Schaffer, B.S. 2010. Does social identity theory underlie relational demography? A test of the moderating effects of uncertainty reduction and status enhancement on similarity effects. Human Relations, 63(7): 903-926
Greene, J.D. & Haidt, J. 2002. How (and where) does moral judgment work? Trends in Cognitive Sciences, 6, 12: 517-523
Greene, J.D., Sommerville, R.B., Nystrom, L.E., Darley, J.M. & Cohen, J.D. 2001. An fMRI investigation of emotional engagement in moral judgment. Sciences, 293: 2105-2108
Greer, L.L. & Van Kleef, G.A. 2010. Equality versus differentiation: The effects of power dispersion on group interaction. Journal of Applied Psychology, online publication:
sept. 6
Haidt, J. & Graham, J. 2007. When morality opposes justice: conservatives have moral intuitions that liberals may not recognize. Social Justice Research. 20: 98-116
Handgraaf, M.J.J., Van Dijk, E., Vermunt, R.C., Wilke, H.A.M., De Dreu, C.K.W. 2008. Less power or powerless? Egocentric empathy gaps and the irony of having little versus no power in social decision making. Journal of Personality and Social Psychology. 95(5):
1136-1149
Hauser, M. D. 2006. Moral minds: How nature designed our universal sense of right and wrong. New York: HarperCollins
Keltner, D., Grunfeld, D., & Anderson, C. 2003. Power, approach and inhibition.
Psychological Review, 110: 265–284
Klinger E., Albaum A. & Hetherington, M. 1964. Factors influencing the severity of moral judgments. The Journal of Social Psychology. 63: 319-326
Kohlberg, L. 1984. The psychology on moral development: Essays on moral development.
San Fransisco: Harper & Row
Kretchmar, J. 2008. Cognitive dissonance theory. Ebcohost Research Starters.
Langer, E. 1975. The illusion of control. Journal of Personality and Social Psychology, 32:
311–328
Lammers, J., Stapel, D.A. & Galinsky, A.D. 2010. Power Increases Hypocrisy. Moralizing in Reasoning, Immorality in Behavior. Psychological Science. 21: 737-744
Lammers, J., Stoker, J.I. & Stapel, D.A. 2009. Differentiating social and personal power:
opposite effects on stereotyping, but parallel effects on behavioral approach tendencies.
Association for Psychological Science, 20(12): 1543-1549
Loewenstein, G.F., Weber, E.U., Hsee, C.K., & Welch, N. 2001. Risk as feelings. Psychology Bulletin, 127: 267–286
Maccoby, E.E. & Wilson, W.C. 1957. Identification and observational learning from films.
Journal of Abnormal Social Psychology, 55: 76-87
Mikhail, J. 2007. Universal moral grammar: Theory, evidence, and the future. Trends in Cognitive Sciences, 11: 143–152
Moghaddam, F.M. & Stringer, P. 1988. Out-group similarity and intergroup bias. The Journal of Social Psychology, 128(1): 105-115
Mulder, M. 1977. The Daily Power Game. Leiden: Martinus Nijhoff
Mullen, B., Brown, R. & Smith, C. 1992. Ingroup bias as a function of salience, relevance, and status: An integration. European Journal of Social Psychology, 22(2): 103-122
Narvaez, D. & Hill, P.L. 2010. The relation of multicultural experiences to moral judgment and mindsets. Journal of Diversity in Higher Education, 3(1): 43-55
Nesdale, D., Durkin, K., Maass, A., Kiesner, J., Griffiths, J., Daly, J., McKenzie, D. 2010. Peer group rejection and children's out-group prejudice. Journal of Applied Developmental Psychology, 31(2): 134-144
Pizarro, D. 2000. Nothing more than feelings? The role of emotions in moral judgment.
Journal for the Theory of Social Behavior, 30(4): 355-375
Reed, A. & Aquino, K.F. 2003. Moral identity and the expanding circle of moral regard toward out-groups. Journal of Personality and Social Psychology, 84(6): 1270-1286
Rest, J.R. 1986. Moral development: advances in research and theory. New York: Preager
Riordan C.M., Schaffer B.S. & Steward M.M. 2005. Relational demography within groups:
Through the lens of discrimination. Dipboye R.L. & Collela, A. (Eds.), Discrimination at work: The psychological and organizational bases: 37–61. Mahwah, NJ: Lawrence Erlbaum Associates
Rodin, J., & Langer, E. 1980. Aging labels: the decline of control and the fall of self-esteem.
Journal of Social Issues, 36: 12–29
Salancik, G. R., & Pfeffer, J. 1974. The bases and use of power in organizational decision making, the case of a university. Administrative Science Quarterly, 19: 453–473
Sivanathan, N., Pillutla, M.M., Murnighan, J.K. 2008. Power gained, power lost.
Organizational Behavior and Human Decision Processes, 105: 135–146
Tajfel, H. & Turner, J. C. 1979. Integrative theory of intergroup conflict. Austin, W.G. &
Worchel, S. (Eds.), The Social Psychology of Intergroup Relations: 33–48. Monterey, CA: Brooks/Cole.
Tajfel, H. & Turner, J. C. 1986. The social identity theory of intergroup behaviour. Worchel S.
& Austin, W.G. (Eds.) Psychology of Intergroup Relations, 2nd edn: 7–24. Chicago, IL: Nelson-Hall
Thompson, S.C. & Armstrong, W. 1998. Illusions of control, underestimation, and accuracy: A control heuristic explanation. Psychological Bulletin, 123: 143–161
Trevino. L. K. 1986. Ethical decision making in organizations: A person-situation interactionist model. Academy of Management Reriew, 11: 601 -617
Tsui, A.S., Egan, T.D. & O’Reilly, C. 1992. Being different: Relational demography and organizational attachment. Administrative Science Quarterly, 37: 549–579
Valdesolo, P. & DeSteno, D. 2007. Moral hypocrisy: social groups and the flexibility of virtue.
Psychological Science, 18(8): 689-690
Wayment, H.A. 2004. It could have been me: vicarious victims and disaster-focused distress.
Personality and Social Psychology Bulletin, 30(4): 515-528
Wrong, D.H. 1968. Some problems in defining social power. American Journal of Sociology, 73: 673–681
