The effects of power on decision deferral

(1)

The effects of power on decision deferral

When, why and how does power influence decision deferral

Master Thesis, Msc Human Resource Management University of Groningen, Faculty of Economics and Business

(2)

2

ABSTRACT

People in power often make important decisions. Previous research found that powerful people have a tendency to act, which suggests that powerful people are less likely to postpone decisions. This study aimed to answer the question when and why powerful people would postpone decisions less than powerless people. It was expected that there would be a negative relationship between power and deferral, in particular when choices are difficult because options are close in attractiveness (low distinctiveness). Furthermore, it was predicted that either subjective experience of certainty (with high power people feeling more certain, even when distinctiveness is low) would mediate the effects of power on choice deferral, only if distinctiveness was low. Finally, it was expected that certainty threshold (i.e., the required level of certainty) would mediate these effects. In a scenario study, in which participants were asked to make six decisions, it was found that distinctiveness yielded several effects. As expected, choices were deferred more often and the subjective experience of certainty was lower when distinctiveness was low. Unexpectedly, results also showed that the certainty threshold was lower, decisions were made less deliberate and participants were more likely to take action when distinctiveness was low. However, contrary to expectations, no effects of power were found. In exploratory analyses it was found that the personality trait impulsiveness moderated the effects of power on some dependent variables: impulsiveness had a negative effect on choice deferral for high distinctiveness choices, and it decreased the certainty threshold and deliberate decision making for low distinctiveness choices Concluding, these findings indicate that distinctiveness has great effects on decision making. Additionally, unlike what was found in power research until now, the experience of power may not always increase action tendencies, and power may interact with personality to affect behavior in some situations.

(3)

3

INTRODUCTION

“If you were given only one choice: To choose or not to choose, which would you choose?” (Helmetter, 1990, p. 53). Choices, options, decisions, alternatives; they are what life is all about. Even if you conclude that none of the alternatives fit your needs, you still make a decision: “When you have to make a choice and don’t make it, that in itself is a choice” (White & Hoffrage, 2009, p. 280). Not choosing is called decision avoidance, or decision deferral when the choice is postponed.

Decision avoidance is a frequently occurring phenomenon, inducing high personal and societal cost (Anderson, 2003). Previous research on decision making focused mainly on why people make certain decisions in different situations and what these decisions are. More recently the ‘how’ of decision making and decision deferral have been studied. How do individuals decide what to choose and when do they refrain from making a choice? In other words, when does a decision become difficult?

(4)

4

Galinsky, Gruenfeld and Magee (2003) studied the connection between the variables power and action. Their outcomes show a positive relationship between power and taking action. Actual or perceived power makes individuals prone to taking action, whereas a lack of power makes individuals refrain from taking action. Decision deferral can be seen as a way to refrain from action, therefore it follows that power has a negative relation with decision deferral. However, this relation may only be found when distinctiveness is low rather than high. Under high distinctiveness decision deferral will not be very prevalent among both powerful and powerless individuals, because one option is clearly superior to other options and there is low uncertainty about which option is the best. However, when distinctiveness is low, I expect that the inclination towards action will lead powerful people to choose, whereas powerless people will more often defer the decision. Therefore I propose there is a negative relationship between power and decision deferral, and that this relation is stronger when distinctiveness is low rather than high.

(5)

5

THEORY

Power is a ubiquitous phenomenon: it exists everywhere; in formal as well as informal settings. Even though it is such a common phenomenon, it is difficult to find one comprehensive definition of power. When first researched, definitions of power focused on the control over valuable resources: “Power has often been construed as the control over money, information or decision making premises” (Anderson, John, & Keltner, 2012, pp. 313-314). Galinsky et al. (2003) also define power according to the control over resources; “Power is the ability to control resources, own and others’, without social interference” (p. 454). Others say power is a fundamental feature of a social organization, creating stability and order (Sligte, Dreu, & Nijstad, 2011). Additionally, Anderson, Keltner and Gruenfeld (2003) define power as “an individual’s relative capacity to modify others’ states by providing or withholding social and/or material resources” (p. 265). The value of these resources or punishments determines the dependence of other individuals on these resources. Their explanation of power corresponds with that of Emerson (1962), who addresses the fact that power to date is perceived as a static phenomenon; “The flaw is the implicit treatment of power as though it were an attribute of a person or group” (p. 31). In his research he highlights the simple fact that without dependency, there is no power. Power is an attribute of the relationship between individuals. We can also look at this definition the other way around like Galinsky, Magee, Gruenfeld, Whitson and Liljenquist (2008) do: “Power, it could also be said, is the capacity to be uninfluenced by others” (p. 1451). Furthermore, power can also be a psychological state – the perception of an individuals’ ability to exert influence over others/situations (Anderson, John, & Keltner, 2012).

For this research I define power as the capacity of one individual to influence others by providing or withholding social and/or material resources that are valuable to another individual.

(6)

6

disinhibited, state/trait driven behavior (Keltner, Gruenfeld, & Anderson, 2003). Disinhibited behavior in this sense, can be explained as being unrestrained, and following one’s true and inner feelings. Low power individuals, in contrast, tend to pay more attention to threats, experience negative emotions and show inhibited, situationally constrained behavior (Keltner, Gruenfeld, & Anderson, 2003).

It is possible that a high and a low power individual in the same situation experience two completely different worlds through their perceptions of reality. The reason behind the fact that powerful individuals experience these elevated levels of affect, cognition and behavior, is the reward-rich environment in which they are (or at least feel) unconstrained by evaluation of others (Keltner, Gruenfeld, & Anderson, 2003). Keltner, Gruenfeld and Anderson (2003) partly based their reasoning on the promotion/prevention focus research by Higgins (1997). This research states that individuals in a promotion focus state concentrate on advancement and accomplishment by attaining their “ideals”: hopes, goals and aspirations. A prevention focus state leads an individual to seek security, protection and safety. These individuals mainly want to avoid mistakes and are likely to take their responsibilities and fulfill their requirements (Shah & Higgings, 2001). Thus, power influences the balance between an individuals’ tendency toward either approach – promotion focus, or inhibition – prevention focus (Keltner, Gruenfeld, & Anderson, 2003). In this light, power seems a positive phenomenon leading to goal achievement and accomplishments of tasks. However, Galinsky et al. (2006) also found that the powerful pay less attention to individuals around them. Through several tests on perspective taking, they found that powerful people view things from their own position, disregarding others’ positions. Powerless people in contrast, are focused on the others’ view. An example of one experiment they conducted is the situation in which participants are asked to draw the letter E on their forehead. High power individuals were three times more likely to draw an E in such a way they would be able to read it themselves – as if they would write it on a piece of paper. For the person sitting opposite, the letter would be mirror inverted. By doing so high power individuals completely disregarded the other person’s point of view.

(7)

7

leads individuals to act. “Freed from social interference, thinking heuristically, experiencing approach- related emotions and insensitive to lurking punishments and threats, the powerful are prone to act” (Galinsky, Gruenfeld, & Magee, 2003, p. 454).

Based on three different experiments, in which power was primed, the results show that power leads directly to action. For instance in one situation, participants were exposed to an annoying fan. They were primed with power by writing a short essay on either a situation in which they were powerful, powerless, or a random situation (control group). The high power individuals turned out to be twice as likely to take action against the annoying fan instead of ignoring it, even though it was not clear whether they were allowed to do so. The general conclusion of this research states that the relation between power and action is ultimately functional: exercising power through action leads to successful goal achievement and eventually, power retention. This power – action relationship can be explained by/results from an environment in which taking action for those with power is taught, conditioned and encouraged. In a surrounding/culture where this specific behavior is rewarded, a natural and positive association between power and taking action will develop (Galinsky, Gruenfeld, & Magee, 2003).

It is likely that power also influences decision making in general and decision deferral in particular, and that powerful individuals with their proclivity towards action less often defer decisions. With many decisions the opportunity exists to choose none of the alternatives. When there are more alternatives available than the ones currently construed, and/or when it is not necessary to take a decision, the option of no-choice − decision deferral − becomes available. Deferral is most likely to occur when the decision is difficult and when the consequences of the decision are significant. An additional cause of decision deferral is regret. “When regret is anticipated, people may try to avoid experiencing it by deferring choice, and this is particularly likely when there is uncertainty about which option is the best” (White & Hoffrage, 2009).

(8)

8

determines how certain an individual needs to be that his preferred alternative is the best option. It can be seen as a minimum standard of certainty to make a decision. Is this minimum standard not reached, than decision deferral could occur.

Since powerful individuals prefer to act rather than defer, power has a negative relationship with deferral. This negative relationship between power and deferral is enhanced when distinctiveness is low. Under high distinctiveness one option is clearly superior to other options and there is low uncertainty about which option is the best. Therefore decision deferral will not be very prevalent among both powerful and powerless individuals in the case of high distinctiveness. However, when distinctiveness is low, I expect that the preference to act will lead powerful people to choose, whereas powerless people will more often defer the decision. This leads to the first hypothesis in this research:

H1: There is a negative relationship between power and deferral, which is enhanced by

distinctiveness.

(9)

9

First, one individual may experience more certainty than another, depending on the situation and personal characteristics. The possession of power leads behavior to change. According to Fast et al. (2011) people with high power are sensitive to rewards, pay more attention to positive and rewarding information, express themselves more freely, and have an inclination towards taking action. Recent insights indicate that powerful individuals’ actions are better aligned with their internal state, or inner feelings, than the actions of powerless individuals (Wisse & Rus, 2012). High distinctiveness increases the subjective experience of certainty. Both high and low power individuals are likely to feel certain about what the better alternative is, since differences are evident. However, only high power people feel certain about their options when distinctiveness is low; their subjective experience of certainty is independent of the distinctiveness between alternatives. Even when distinctiveness is low, people with power tend to listen to their gut feelings, trust these inner feelings, and base their actions on these feelings (Wisse & Rus, 2012). Individuals high in power are inclined to take action, leading to less deferral. This relationship can be explained by the fact that they have a higher subjective experience of certainty, even in situations where distinctiveness is low. This leads to hypothesis H2a.

H2a: Subjective experience of certainty mediates the effects of power on choice deferral, only if

distinctiveness is low.

(10)

10

there is “a natural fit between promotion focus concerns and the use of eagerness approach means because eagerness means ensure the presence of positive outcomes like ensuring hits and means of advancement” (p. 179). This eagerness approach for positive outcomes leads individuals to ensure themselves against omission errors – incorrectly refraining from action (Leach & Plaks, 2009). Individuals high in power are inclined to act (Galinsky, Gruenfeld, & Magee, 2003). According to the promotion focus theory (Higgins, 2002) high power individuals prefer to act to achieve positive outcomes and make sure they don’t miss out on an attractive alternative. They rather choose with the risk of picking a moderately attractive alternative than waste the best alternative. In other words, individuals high in power do not need to be as certain as individuals low in power when it comes to choosing the best alternative, they simply want to prevent walking away empty-handed. For them, an error of omission is more severe than the error of commission; “nothing ventured – nothing gained” (Leach & Plaks, 2009, p. 221).

These phenomena lead the certainty threshold to be lower for high power people than for low power people, which in turn would decrease deferral. For high power individuals the difference in attractiveness between alternatives would not have to be as high as it would have to be for low power individuals (lower distinctiveness). In sum, high power individuals are less likely to defer. This can be explained by the fact that their threshold to be certain is lower; they care more about making a decision and less about the outcome, even in the case distinctiveness is low. This leads to hypothesis H2b.

(11)

11

METHOD

Participants and design

Participants were 148 students (74 males and 73 females, one participant did not indicate gender) of the University of Groningen who participated individually. For their participation respondents had the option of either receiving research credit points or a financial compensation of €7. The average age of the participants was 20.98 years (SD = 2.26).

The experiment consisted of a mixed design. There were three between subjects conditions for power; a high power condition, a low power condition and a control condition. Participants were randomly assigned to these conditions. Additionally experienced distinctiveness was manipulated within participants by means of six decisions that were high vs. low on distinctiveness (three high vs. three low distinctiveness for every participant), presented in random order. All six decisions had both a low and a high distinctiveness version. In version one, decisions 1, 2, and 3 were high in distinctiveness whereas decisions 4, 5, and 6 were low in distinctiveness. This was reversed for version two where decisions 1, 2, and 3 were low in distinctiveness and decisions 4, 5, and 6 were high in distinctiveness. For example, decision one had a low distinctiveness version in which alternatives were approximately equally attractive, making it likely that the decision is perceived as difficult by the participant. On the other hand there was a high distinctiveness version of decision one in which alternatives were clearly different in attractiveness, making it easy for the participant to choose the better alternative.

Procedure

(12)

12

had power over another person, in which someone had power over them or for the control condition what they ate for breakfast the previous day. Participants were shown any one of the following texts (depending on their condition):

1) High power condition

For this second part of the study, we would like you to recall and write about a personal incident in which you had power over another individual or individuals. In the box below you can briefly describe this memory in 5 to 10 lines. It can be any situation from your personal or professional life in which you controlled the ability of another person or persons to get something they wanted or when you were in a position to evaluate those individuals.

2) Low power condition

For this second part of the study, we would like you to recall and write about a personal incident in which someone else had power over you. In the box below you can briefly describe this memory in 5 to 10 lines. It can be any situation from your personal or professional life in which someone else controlled your ability to get something you wanted, or who was in a position to evaluate you.

3) Control condition

For this second part of the study, we would like you to recall and write about what you had for breakfast yesterday. In the box below you can briefly describe this memory in 2 to 5 lines.

(13)

13

decisions 1, 2, and 3 were high distinctiveness and decisions 4, 5, and 6 were low distinctiveness. In version two decisions 1, 2 and 3 were low distinctiveness and decisions 4, 5 and 6 were high distinctiveness (see table 1).

Table 1; distribution of decisions, high vs. low distinctiveness

Version number Distinctiveness Decision number

1 High 1, 2, and 3

1 Low 4, 5, and 6

2 High 4, 5, and 6

2 Low 1, 2, and 3

Participants were asked to imagine that they were thinking of making a purchase in the categories presented (digital camera, laptop, citytrip, restaurant, mobile phone and apartment). It was emphasized beforehand that there were no right or wrong answers and that the participant should choose the option that best reflects his/her preferences. As in reality, participants also had the option to choose none of the options provided and look for different alternatives elsewhere (the deferral option). Accompanying every decision, participants were asked to answer a few questions about their decision. An example decision is:

Imagine that you would like to buy a digital camera. At the store you normally shop, you find the following two alternatives. You also have the option of not buying either of them and look for different alternatives elsewhere.

Low distinctiveness version

Digital camera 1 Digital camera 2

Features:

- 16 pixels

- 30 fps (frames per second) - 6x optical zoom

- Internal and external memory - Wi-Fi - Bluetooth - Black € 125 Features: - 16 pixels

- Internal and external memory - Pictbridge

- HD- connection - Black

(14)

14 High distinctiveness version

Features:

- 16 pixels

- External memory - Black

€ 150

The task the participant had, was to decide to either choose option A, option B, or none of the above and look for different alternatives elsewhere, option C (deferral). To measure the different variables (e.g. subjective experience of certainty, certainty threshold), the next step in the procedure was to present the participants with questions for each decision on distinctiveness, experienced certainty, the certainty threshold and some general questions measuring how deliberate or impulsive the decision was made.

Measures

The first personality questionnaire was based on 12 items on “maximizing” (Schwartz, Ward, Monterosso, Lyubomirsky, White, & Lehman, 2002), which were measured on a 5- point Likert scale (completely disagree – completely agree). Example questions are: 1) When I watch TV, I often change channels, even when I try to watch one show. 2) I often fantasize about leading a completely different life than my current life. 3) I never accept second-best. Cronbachs α for the maximizing scale was .60. Since this is not a reliable score, this data was not used in the final analysis.

The second personality questionnaire consisted of 7 items on “impulsiveness” (Patton, Stanford, & Barret, 1995), again rated on a 5- point Likert scale (completely agree – completely disagree). Some sample questions are: 1) I act “on impulse”, 2) I do things without thinking, and 3) I am happy-go-lucky. After scaling these items into one variable, Cronbachs α was .71.

(15)

15

items measure to what extent a person is agreeable, conscientious, neurotic, extraverted and open. Example questions for each type are 1) (agreeableness) As a person I am Sympathetic, 2) (conscientiousness) As a person I am Efficient, 3) (neuroticism) As a person I am Nervous, 4) (extraversion) As a person I am Assertive and 5) (openness) As a person I have Wide interests. All 25 items were scaled and divided into the 5 measured groups; Agreeablenes (α .68), Conscientiousness (α .74), Neuroticism (α .78), Extraversion (α .73) and Openness (α .65).

After every decision participants were asked to make a decision between either option A, option B, or option C; none of the above and look for different alternatives elsewhere (deferral). This way the number of deferrals in high and low distinctiveness situations could be measured by adding the number of deferrals for each low- and high distinctiveness decision. Scores ranged from minimum o to maximum 3.

After this decision, participants were asked to answer questions concerning the variables distinctiveness, experienced certainty, the certainty threshold and some general questions on how deliberate/impulsive the decision was made. The questions concerning distinctiveness are 1) How attractive did you find X (camera, laptop etc,) one?, 2) How attractive did you find X two?, 3) The two X’s were approximately equally attractive. And 4) It was clear what the best of the two options was. For all six decisions, the absolute difference between items 1 and 2 was calculated. Item 3 had to be recoded. For all 4 items Z- scores were computed and they were scaled into 1 variable. Cronbachs Alpha scores for all decisions ranged from α .86 to α .90. The specific scores can be found under “experienced distinctiveness” in table 2.

To measure experienced certainty, three questions per decision were asked: 1) I am certain about my decision. 2) I feel confident about my decision. And 3) I am certain which X was the best. For all six decisions these three items were scaled, Cronbachs Alpha scores for all decisions ranged from α .76 to α .88. The specific scores per decision can be found in table 2 under “experienced certainty”.

(16)

16

available alternative. Again for all six decisions these three items were scaled, Cronbachs Alpha scores for all decisions ranged from α .74 to α .82 and specific scores can be found in table 2 under “certainty threshold”.

Finally, nine general questions per decision were asked. Participants were asked to what extent the following words described the way they made their decision: 1) Deliberate, 2) Well thought through, 3) Being careful, 4) Avoid mistakes, 5) Choosing the best X, 6) Impulsive, 7) Spontaneous, 8) Taking action, 9) Choosing an X that was good enough. These items were separated into two variables; deliberate (items 1 – 5) and taking action/impulsiveness (items 6 - 9) based on theoretical and empirical reasoning. For the variable taking action/impulsiveness, the last item (choosing an X that was good enough) lowered the Cronbachs Alpha score, and was therefore left out of the analysis. Cronbachs Alpha scores for the variable deliberate ranged from α .64 to .75. For the variable taking action/impulsiveness (without the last item; choosing an X that was good enough) ranged from α .61 to .71. Separate Cronbachs Alpha scores for each decision on these two variables can be found in table 2 under “deliberate” and “taking action/impulsiveness”.

Table 2; Cronbachs Alpha scores

Measures Decision 1 2 3 4 5 6 Experienced distinctiveness .90 .86 .88 .88 .89 .87 Experienced certainty .85 .81 .88 .85 .76 .77 Certainty Threshold .82 .76 .79 .80 .74 .77 Deliberate .74 .71 .70 .70 .64 .75 Taking action/impulsiveness .70 .65 .61 .71 .69 .67

An overview of all decisions (high and low distinctiveness versions) and questions used in this research can be found in appendix 1.

(17)

17

RESULTS

Manipulation checks

To check the manipulation of distinctiveness, a mixed-model ANOVA was performed. In this ANOVA the experienced distinctiveness of high versus low distinctiveness choices was the within-subjects variable, and power and version were between-subjects variables. This analysis showed the expected main effect of the within-subjects variable distinctiveness, F (1, 142) = 566.54, p < .001, experienced distinctiveness was higher for high distinctiveness (M = .60, SD = .44) than for low distinctiveness choices (M = -.60, SD = .42), confirming that the distinctiveness manipulation was successful. The within-subject variable distinctiveness did not interact with the between-subjects variables power or version, all F < 1.10, p > .34, indicating that the distinctiveness manipulation was successful across conditions. There was, however, also a significant version x power interaction. Across high and low distinctiveness choices, experienced distinctiveness in version 1 was higher in the control condition (M = .14) than in the low (M = -.08) and high (M = -.09) power conditions. In version 2, overall experienced distinctiveness was higher in the low power condition (M = .06) than in the high power (M = -.01) or control (M = -.03) conditions. This effect is hard to explain, and not of substantive interest. Additionally the distinctiveness manipulation for every one of the six decisions was checked using crosstabs analysis. This analysis showed that for every decision in the high distinctiveness version the less attractive alternative was never chosen, whereas in the low distinctiveness version the choice of alternatives was spread among options A, B, and C.

(18)

18 Descriptive statistics

(19)

19 Table 3; descriptive statistics all variables

Version one Version two

Measures Low power

High power Control Low power High power Control

Deferral High distinctiveness M = .52

(SD=.77) M = .64 (SD=.91) M = .67 (SD=.76) M = .44 (SD=.58) M =.40 (SD=.58) M = .42 (SD=.58)

Deferral Low distinctiveness M = .80

(SD=.71) M = .92 (SD=.91) M = .79 (SD =.78) M =1.20 (SD=.96) M =1.00 (SD=1.00) M = 1.04 (SD=.86) Experienced distinctiveness high distinctiveness M =.58 (SD=.41) M = .47 (SD=.50) M = .71 (SD=.38) M = .62 (SD=.42) M = .63 (SD=.42) M =.57 (SD=.51) Experienced distinctiveness low distinctiveness M = -.73 (SD=.41) M = -.66 (SD=.34) M = -.42 (SD=.43) M = -.50 (SD=.40) M = -.65 (SD=.47) M = -.61 (SD=.39)

Experienced certainty high distinctiveness M = 4.46 (SD=.42) M = 4.43 (SD=.48) M = 4.47 (SD=.53) M = 4.46 (SD=.41) M = 4.44 (SD=.46) M = 4.37 (SD=.56)

Experienced certainty low distinctiveness M = 3.64 (SD=.46) M = 3.51 (SD=.43) M = 3.78 (SD=.60) M = 3.71 (SD=.53) M = 3.41 (SD=.76) M = 3.45 (SD=.62)

Certainty threshold high distinctiveness M = 3.77 (SD=.82) M = 3.98 (SD=.63) M = 3.99 (SD=.55) M = 3.66 (SD=.84) M = 3.90 (SD=.57) M = 3.79 (SD=.69)

Certainty threshold low distinctiveness M = 3.26 (SD=.60) M = 3.68 (SD=.63) M = 3.57 (SD=.64) M = 3.58 (SD=.83) M = 3.70 (SD=.58) M = 3.42 (SD=.60) Deliberate high distinctiveness M = 3.63 (SD=.40) M = 3.69 (SD=.59) M = 3.83 (SD=.43) M = 3.54 (SD=.55) M = 3.55 (SD=.49) M = 3.63 (SD=.52) Deliberate low distinctiveness M = 3.40 (SD=.43) M = 3.57 (SD=.49) M = 3.51 (SD=.48) M = 3.41 (SD=.67) M = 3.38 (SD=.45) M = 3.25 (SD=.48)

Taking action high distinctiveness M = 2.98 (SD=.67) M = 2.90 (SD=.67) M = 3.06 (SD=.68) M = 3.07 (SD=.80) M = 3.05 (SD=.57) M = 2.92 (SD=.64)

(20)

Table 4; correlations table Correlations Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 Impulsiveness 1 2 Agreeableness .09 1 3 Conscientiousness -.29*** _.08 ₁ 4 Neuroticism -.01 -.01 -.04 1 5 Extraversion -.01 .31*** .35*** -.00 1 6 Openness .13 .34*** _.05 _-.09 _.43*** ₁

7 Deferral High Distinctiveness -.11 -.11 .05 -.10 -.02 .04 1

8 Deferral Low Distinctiveness -.13 -.13 .00 -.13 .02 .06 .25** ₁

9 Distinctiveness, high distinctiveness .03 -.03 -.07 .03 .09 .07 -.26** _.02 ₁ 10 Distinctiveness, low distinctiveness .13 .07 -.00 .16 .19* _.12 _.08 _-.15 _.01 ₁ 11 Experienced Certainty High Distinctiveness .10 .12 -.07 -.02 .08 .09 -.23** _-.15 _.59*** _-.05 ₁ 12 Experienced Certainty Low Distinctiveness .03 .08 .07 .09 .21* _.25** _.11 _-.16* _-.16 _.56*** _.11 ₁ 13 Threshold High Distinctiveness -.12 .07 .07 .15 .11 .09 .22** _.01 _.27** _.08 _.32*** _.04 ₁ 14 Threshold Low Distinctiveness -.21* _.09 _.07 _.20* _.17* _.10 _.23** _.28** _.06 _.28** _.08 _.26** _.59*** ₁ 15 Deliberate Decision- making High Distinctiveness -.10 .27** _-.04 _.14 _.10 _.12 _.16 _-.22** _.19* _.07 _.23** _.02 _.54*** _.35*** ₁ 16 Deliberate Decision- making Low Distinctiveness -.17* _.22** _.03 _.10 _.15 _.20* _.19* _.22** _.06 _.10 _.03 _.15 _.32*** _.56*** _.54*** ₁ 17 Taking Action High Distinctiveness .49*** _.05 _-.07 _.03 _-.03 _.11 _-.24** _-.14 _.01 _.04 _.12 _.10 _-.21** _-.26** _-.24** _-.08 ₁

18 Taking Action Low Distinctiveness .36*** _.15 _-.08 _.10 _.07 _.12 _-.18* _-.49*** _-.01 _.13 _.23** _.19* _-.00 _-.14 _.11 _-.17* _.58*** ₁ *** Correlation is significant at the 0.001 level (2-tailed)

(21)

21 Hypothesis 1

Hypothesis 1 predicted that the number of deferrals would be lower in the high power condition as compared to the low power and control condition, and that this would be the case especially for low distinctiveness choices. To examine this hypothesis, a mixed-model ANOVA was performed, in which distinctiveness (of high vs. low distinctiveness choices) was the within-subjects variable, and power and version were between subjects variables. This analysis yielded a main effect of distinctiveness, F (1, 142) = 31.31, p < .001, showing that more choices were deferred when choices were low in distinctiveness (M = 0.96, SD = 0.87) than when choices were high in distinctiveness (M = 0.51, SD = 0.71). There was also an unexpected distinctiveness x version interaction, F (1, 142) = 7.42, p = .007, showing that the effect of distinctiveness on deferral was less strong in version 1 (high distinctiveness M = 0.61; low distinctiveness M = .84) than in version 2 (high distinctiveness M = 0.42; low distinctiveness M = 1.08). However, the predicted distinctiveness x power interaction was not significant, F (2, 142) = 0.28, p > .76. This results implies that Hypothesis 1 must be rejected: although high distinctiveness choices were deferred less often than low distinctiveness choices, this effect was not moderated by power. No further effects were significant, all F < 1.

Hypothesis 2 A

(22)

22 Hypothesis 2 B

The same test was performed to test the final hypothesis H2b. This hypothesis predicted that certainty threshold explained lower number of deferrals in the high power condition as opposed to low power or control, but only if distinctiveness is low. Again a mixed-model ANOVA was conducted to test this hypothesis. Certainty threshold for high and low distinctiveness choices was the within-subjects variable. Power and version remained the between subjects variables. The analysis yielded a main effect of distinctiveness, F (1, 142) = 40.07, p < .001, showing that the certainty threshold was higher for high distinctiveness choices (M = 3.85, SD = 0.69) than it was when choices were low in distinctiveness (M = 3.54, SD = 0.66). No further effects were significant, all F < 3.8, p > 0.05. The results show that Hypothesis 2 B has to be rejected: the certainty threshold is lower when distinctiveness is lower, and no effects of power were found.

Additional results

Additionally participants were tested on how they made their decisions. This led to two variables per decision: deliberate decision making and taking action. For both variables a within-subjects ANOVA was conducted to examine effects of power. For the first test deliberate decision making in the high and low distinctiveness situation was the within subjects factor and, as with the other tests, power and version were the between subject factors. Only a significant effect for distinctiveness was found, F (1, 142) = 32.65, p < .001. The decision was made more deliberate when distinctiveness was high (M = 3.64, SD = .50) than when distinctiveness was low (M = 3.42, SD = .51). Additionally there was an unexpected between-subjects effect for the version F (1,142 = 4.07, p = 0.046. Deliberate decision making was less strong across high and low distinctiveness in version 1 (total version 1 M = 3,61) than in version 2 (total version 2 M = 3,46). No significant effects of power were found.

(23)

23

So far, no effects of power have been found. It is possible, however, that power and personality interact to affect decision deferral. From the correlations it appears that of the personality variables impulsiveness shows the strongest relations with the dependent variables, and I therefore decided to explore the effects of impulsiveness further. When assessing the outcomes of this part of the research, it is important to keep in mind that these effects were not hypothesized; they are merely exploratory analysis to check for any additional outcomes. It is possible, for example, that power reduces decision deferral especially when impulsiveness is high and for low distinctiveness choices.

(24)

24

Figure 1; graphic presentation linear regression analysis deferral

The same analysis was performed in which deferral for low distinctiveness choices was the dependent variable. No significant effects were found; all p > 0,05.

(25)

25

Figure 2; graphic presentation linear regression analysis deliberate decision making

(26)

26

Figure 3; graphic presentation linear regression analysis threshold

Table 5; linear regression exploratory research

Dependent variable Independent variable Deferral, high distinctiveness Deliberate decision making, low distinctiveness

Certainty Threshold, low distinctiveness

Low Power Dummy B -0.02 (-0.16) B 0.06 (0.60) B 0.04 (-0.27) High Power Dummy B -0.03 (-0.23) B 0.10 (0.98) B 0.19 (1.43) Z- score Impulsiveness B 0.13 (1.29) B 0.01 (0.18) B 0.01 (0.15) Low Power x Impulsiveness B -0.48 (-3.38)** B -0.27 (-2.66)** B -0.32 (-2.43)* High Power x Impulsiveness B -0.15 (-1.08) B -0.04 (-0.38) B -0.13 (-1.00)

- scores 0.09* 0.09* 0.11**

(27)

27

DISCUSSION

Summary of Results

Many studies have been performed on both the concept of power as well as decision making. The aim of this research was to combine these two aspects to find out if, when, why and how power influences the process of decision making, and decision deferral in particular. Three hypotheses were proposed involving the relationship between power and decision deferral, experienced distinctiveness of alternatives, experienced certainty and certainty threshold. I argued that power would lead to less decision deferral if distinctiveness was low (hypothesis 1), and that this negative relationship is mediated by either the subjective experience of certainty (hypothesis 2A), or by certainty threshold (hypothesis 2B). These ideas were tested in a controlled lab setting at the University of Groningen. Participants were students who all participated individually. The experiment consisted of a mixed design with three between subjects conditions for power – high power, low power and control condition – to which participants were randomly assigned. Furthermore, experienced distinctiveness was manipulated within participants providing them with six decisions on either high or low distinctiveness (three high vs. three low, randomly assigned).

The main findings from this study confirm effects of distinctiveness on decision deferral. On all measures, there was a significant effect for distinctiveness. It seems that low distinctiveness did not only reduce certainty and increase decision deferral, as was found in previous research (Nijstad, 2011), it also changed decision making strategies. When distinctiveness was low, decisions were made less deliberate, participants were more prone to taking action and their certainty threshold decreased. From logical reasoning, I would predict these outcomes to be the other way around; as it becomes more difficult to assess which alternative is the best, I would expect the certainty threshold to be higher, as well as the time to think things over to be longer. I would also expect participants to refrain from action when it is more difficult to select the best alternative. Current research does not give a clear explanation for these results. Perhaps when it is more difficult to assess what the better alternative is, participants start to care less about the outcome of their choice – lowering the threshold, making them less deliberate and just take action to be done with it.

(28)

28

manipulation did not work as intended. However, the same manipulation has been used before by Galinsky, Gruenfeld and Magee (2003), and participants have generally complied with instructions very well. It is also possible that the manipulation exercise did attain the right mindset, but effects wore of before all decisions had been finished. A second possibility is the nature of the decisions. Participants were shown consumer decisions that would not affect themselves in a great way, or others at all. Perhaps when decisions were asked that would have an effect on other individuals, outcomes would have been different. Finally, an explanation could be the fact that participants were students who possibly had difficulty relating to high- and low power situations. These situations might become more evident during the working- life stage in which high- and low power differences become more clear and have bigger impact.

From the exploratory analysis, effects of impulsiveness on deferral, deliberate decision making and certainty threshold were found. This in itself is not strange: it can be expected that impulsive individuals would choose rather than defer, have a lower certainty threshold and make a decision less deliberate, since impulsive individuals tend to act before they think. Unexpected was that all significant effects were only found in the low power condition. Participants that were high in impulsiveness and in the low power condition, were less likely to defer for high distinctiveness choices, and made decisions less deliberate and had a lower certainty threshold for low distinctiveness choices. A possible explanation could be that impulsive individuals low in power feel that less depends on their decision, making them less likely to defer.

Implications

(29)

29

previously was assumed or less widely applicable. Additionally, the effects might not apply equally to every personality type and/or situation. The effects of personality found in this study – especially of impulsiveness – address a whole new area of research in relation to decision making. Perhaps the suggested effects of power are only present when impulsiveness is high or are moderated by other personality variables.

In practice, it is highly useful to know when, why and how individuals make or defer decisions. For organizations this knowledge could lead to a more effective and efficient (less financial resources and less time consuming) decision making policy. However, known effects of power might not always hold in every decision making situation. Organizations need to be cautious when interpreting effects of power since there might be influences of personality. It would be useful for organizations to take the effects of personality – and impulsiveness in particular – into account, when for instance developing the decision making policy.

Limitations and future directions

(30)

30

different group of participants. Focusing on students might have led to somewhat distorted experiences of high and low power, as well as the familiarity of the participants with making the given decisions. A different/more diverse group of participants, with experience in high and low power situations – for instance in a professional context – may give a more realistic display of the effects for practical implications. Additionally, changing the nature of the decisions could be interesting. Participants were shown consumer decisions with no real consequences. Perhaps when decisions were asked that would have an effect on other individuals, outcomes would have been different. Finally, current theory does not lead to an explicit explanation for the lack of effects of power in relation to decision making. This might be a useful addition in future research; effects of power might interact with personality and/or different situations. We know power changes the behavior of individuals (Keltner et al, 2003; Galinsky, Magee, Greunfeld, Whitson, & Liljenquist, 2008), but effects might differ due to personality traits, effects could be less strong than previously was assumed or they could be less widely applicable.

Conclusion

(31)

31

REFERENCES

Anderson, C. J. (2003). The psychology of doing nothing: Forms of decision avoidance result from reason and emotion. Psychological Bulletin, 1, 139-167.

Anderson, C., John, O. P., & Keltner, D. (2012). The personal sense of power. Journal of Personality, 80, 313-344.

Bendersky, C., Hays, N. A. (2012). Status conflict in groups. Organization Science, 2, 232-340.

Christie, A. M., Barling, J. (2010). Beyond status: Relating status inequality to performance and health in teams. Journal of Applied Psychology, 5, 926-927.

Corbin, R. (1980). Decisions that might not get made. In Wallsten, T.S. (Ed.). Cognitive processes in choice and decision behaviour, (pp. 47-67). Hillsdale, NJ: Erlbaum.

Dhar, R. (1996). The effect of decision strategy on deciding to defer choice. Journal of Behavioral Decision Making, 4, 265-281.

Dhar, R. (1997). Consumer preference for a no-choice option. Journal of Consumer Research, 4, 215-231.

Dhar, R., & Nowlis, S. (1999). The effect of time pressure on consumer choice deferral. Journal of Consumer Research,4, 369-384.

Dhar, R., & Simonson, I. (2003). The effect of forced choice on choice. Journal of Marketing Research, 2, 146-160.

Emerson, R. M. (1962). Power-dependence relations. American Sociological Review, 1, 31-41.

Fast, N., Sivanathan, N., Mayer, N., & Galinsky, A. (2011). Power and overconfident decision-making. Organizational Behavior and Human Decision Processess, 117, 249-260.

Galinsky, A. D., Gruenfeld, D. H., & Magee, J. C. (2003). From power to action. Journal of Personality and Social Psychology, 3, 453-466.

Galinsky, A., Magee, J., Greunfeld, D., Whitson, J., & Liljenquist, K. (2008). Power reduces the press of the situation: Implications for creativity, conformity and dissonance. Journal of Personality and Social Psychology, 6, 1450–1466.

Galinsky, A., Magee, J., Inesi, M., & Gruenfeld, D. (2006). Power and perspectives not taken. Psychological Science, 12, 1068-1074.

(32)

32

Healy, D., Hogg, M., & Terry, D. (2006). It's just not fair: The effects of status and legitimacy on group-based persuasion. In Mavor, Kenneth I., Australian Journal of Psychology, (pp.31-31) London: Taylor & Francis.

Helmetter, S. (1990). Choices: Discover your 100 most important life choices, (pp. 53-53) New York: Pocket Books.

Higgins, E. (2002). How self-regulation creates distinct values: The case of promotion and prevention decision making. Journal of Consumer Psychology, 3, 177-191.

Higgins, E., & Crowe, E. (1997). Regulatory focus and strategic inclinations: Promotion and prevention in decision-making. Organizational behavior and human decision processes, 2, 117–132.

John, O., & Srivastava, S. (1999). The big five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O.P. John (Eds.), Handbook of personality theory and research (pp. 102–138). New York: Guilford Press.

Katzenbach, J. R., & Smith, D. K. (1994). Teams at the top. McKinsey Quarterly, 1, p. 71-79.

Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, Approach, and Inhibition. Psycholgical Review, 2, 265-284.

Kim, S., & McGill, A. L. (2011). Gaming with mr. slot or gaming the slot machine. Power, anthropomorphism, and risk perception. Journal of Consumer Research,1, 94-107.

Leach, F., Plaks, J. (2009). Regret for errors of commission and omission in the distant term vs. near term: The role of level abstraction. Personality and Social Psychology Bulletin, 2, 221-229.

Nijstad, B. (2011). Moeilijke beslissingen. Gedrag & Organisatie, 24, 304-315.

Patton, J., Stanford, M., Barret, E. (1995). Factor structure of the Barratt impulsiveness scale. Journal of Clincal Psychology, 6, 768-774.

Phillips, K. W., Rothbard, N. P., Dumas, T. L. (2009). To disclose or not to disclose? Status distance and self-disclosure in diverse environments. Acadamy of Management Review, 4, 710-732.

Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D. (2002). Maximizing versus satisficing: Happiness is a matter of choice. Journal of Personality and Social Psychology, 83, 1178-1197.

Shah, J., Higgings, E. (2001). Regulatory concerns and appraisal efficiency: The general impact of promotion and prevention. Journal of Personality and Social Psychology, 5, 693-705.

(33)

33

Stoker, J. (2008). Effects of team tenure and leadership in self-managing teams. Personnel Review, 5, 564-582.

White, C. M., & Hoffrage, U. (2009). Testing the tyranny of too much choice against the allure of more choice. Psychology & Marketing, 3, 280-298.

(34)

34

APPENDIX 1

Overview decisions and questions

Personality tests

Maximizing: 1 = completely disagree – 5 = completely agree

When I watch TV, I often change channels, even when I try to watch one show

1 2 3 4 5

When I listen to the radio in my car, I often check other stations to look for a better show, even when I like the show that is on.

1 2 3 4 5

I treat relationships like clothes; I expect to try out allot before finding the right one.

1 2 3 4 5

It doesn’t matter how satisfied I am with my current job, it is always a good thing to keep my eyes open for better

opportunities.

1 2 3 4 5

I often fantasize about leading a completely different life than my current life.

1 2 3 4 5

I am a big fan of lists that rank things (best movies, best singers, best athletes, best books etc.).

1 2 3 4 5

I often find it difficult to buy a gift for a friend. 1 2 3 4 5 Renting a DVD is difficult, I always struggle choosing the best. 1 2 3 4 5 Writing is very difficult, even a letter to a friend, because it is

hard to formulate things the right way. I make several drafts, even for simple things.

1 2 3 4 5

Regardless of what I do, I set the highest standards for myself. 1 2 3 4 5

I never accept second-best. 1 2 3 4 5

Any time a choice is presented to me, I try to imagine all other options, even the ones that are not present at that time.

(35)

35

Impulsiveness: 1 = completely disagree – 5 = completely agree

I act “on impulse” 1 2 3 4 5

I act on the spur of the moment 1 2 3 4 5

I buy things on impulse 1 2 3 4 5

I make up my mind quickly 1 2 3 4 5

I do things without thinking 1 2 3 4 5

I spend more than I earn 1 2 3 4 5

(36)

36

Big 5: 1 = completely disagree – 5 = completely agree 1-5 = agreeableness, 6-10 = conscientiousness, 11-15 = neuroticism, 16-20 = extraversion

(37)

37 Decisions

1. Imagine that you would like to buy a digital camera. At the store you normally shop, you find the following two alternatives. You also have the option of not buying either of them and look for different alternatives elsewhere.

Features:

- 16 pixels

- Internal and external memory - Pictbridge

- HD- connection - Black

€ 120

High distinctiveness version

Features:

- 16 pixels

- External memory - Black

€ 150 In this situation I would (please click your answer below): A Buy camera number 1

B Buy camera number 2

(38)

38

2. Imagine that you would like to buy a laptop. At the store you normally shop, you find the following two alternatives. You also have the option of not buying either of them and look for different alternatives elsewhere.

Laptop 1 Laptop 2

Features:

- 300 GB harddisk - 16 inch screen

- Durability battery 3 hours - DVD- player

- Incl. MS Office package - Black

€ 625

Features:

- Durability battery 3 hours - Integrated webcam - Touchscreen - Silver

€ 625

Laptop 1 Laptop 2

Features:

- Durability battery 2,5 hours - Black

€ 650

Features:

- Durability battery 3 hours - Integrated webcam - Touchscreen - Silver € 625

In this situation I would (please click your answer below): A Buy laptop number 1

B Buy laptop number 2

(39)

39

3. Imagine that you are looking to book a city trip. At the website you normally book your trips, you find the following two alternatives. You also have the option of not booking either of them and look for different alternatives elsewhere.

Citytrip 1 Citytrip 2

Features:

- 1 hour 32 minutes flight

- Easy, quick and cheap transport to the city center

- High probability for sunny weather - Great food/restaurants

- Beach

- Many opportunities for shopping - Many architectural sites

€ 650

Features:

- Many historical sites - Theaters

- Clubs

€ 650

Citytrip 1 Citytrip 2

Features:

- Beach

- Many opportunities for shopping - Many architectural sites

€ 650

Features:

- Difficult, slow and expensive transport to get to the city center - High probability for sunny weather - Great food/restaurants

- Crowded with tourists

€ 750 In this situation I would (please click your answer below): A Book citytrip number 1

B Book citytrip number 2

(40)

40

4. Imagine that you are looking for a nice restaurant in a new city. The trip advisor app you downloaded recommends the following two alternatives. You also have the option of not choosing either of them and look for different alternatives elsewhere.

Restaurant 1 Restaurant 2

Features:

- Good appetizers

- Wide selection of main courses - Fast and friendly service - Plenty parking space nearby - Cozy atmosphere

- Recommended by friends

€ 50

Features:

- Good appetizers

- Wide selection of main courses - Fast and friendly service

- Location is nearby (walking distance) - Good review in the local paper - Well visited

€ 45

Restaurant 1 Restaurant 2

Features:

- Good appetizers

- Wide selection of main courses - Fast and friendly service - Need to drive there by car - Crowded

€ 60

Features:

- Good appetizers

- Wide selection of main courses - Fast and friendly service

- Location is nearby (walking distance) - Good review in the local paper - Well visited

€ 45 In this situation I would (please click your answer below): A Choose to go to restaurant number 1

B Choose to go to restaurant number 2

(41)

41

5. Imagine that you are looking to buy a new mobile phone. At the store you normally shop, you find the following two alternatives. You also have the option of not choosing either of them and look for different alternatives elsewhere.

Mobile phone 1 Mobile phone2

Features:

- Screen size 4.2 inch - 16 GB memory - Double sided camera - Android 4.1.2

- 1 year subscription - € 40 per month

€ 100

Features:

- Screen size 4 inch - 16 GB memory - Double sided camera - Android 4.1.1

- 2 year subscription

- First 9 months 40% discount, than €50 per month

€ 125

Mobile phone 1 Mobile phone2

Features:

- Screen size 4.2 inch - 16 GB memory - Double sided camera - Android 4.1.2

- 1 year subscription - € 40 per month

€ 100

Features:

- Screen size 3.8 inch - 8 GB memory - Single sided camera - Android 4.1.1 - 2 years subscription - €50 per month

€ 250 In this situation I would (please click your answer below): A Buy mobile phone number 1

B Buy mobile phone number 2

(42)

42

6. Imagine that you are looking to rent an apartment. The real- estate agent finds you the following two alternatives. You also have the option of not choosing either of them and look for different alternatives elsewhere.

Apartment 1 Apartment 2

Features: - 80 m2 - New kitchen - New carpeting - Private parking space - Backyard 15 m2

- Fully furnished living room - Security deposit €250 - Suburb location € 575 Features: - 80 m2 - New kitchen - New carpeting - Dishwasher available - Refrigerator available - Fully furnished living room - Security deposit €300 - Central location

€ 500

Apartment 1 Apartment 2

Features: - 80 m2 - New kitchen - New carpeting - Private parking space - Security deposit €250 - Bad neighborhood € 575 Features: - 80 m2 - New kitchen - New carpeting - Dishwasher available - Refrigerator available - Fully furnished living room - Security deposit €300 - Central location

€ 500 In this situation I would (please click your answer below): A Rent apartment number 1

B Rent apartment number 2

(43)

43

After every one of the previous six decisions participants were asked to answer the following questions, keeping the decision they just made in mind.

Distinctiveness:

How attractive did you find X 1 (dist1) How attractive did you find X 2 (dist 2)

The two X were approximately equally attractive. (Dist 3) It was clear what the best of the two options was. (dist 4)

Experienced certainty:

I am certain about my decision. (Excert 1) I feel confident about my decision (excert 2)

I am certain which X was the best option. (excert 3)

Certainty threshold:

I wanted to be absolutely certain I chose the best alternative. (Certhres 1)

I would only choose X 1 or X 2 if I’m certain it is the best available alternative. (certthres 2) It was important to me to choose the best available alternative. (certthres 3)

General 1 - 9 , Deliberate and taking action / impulsiveness

To what extent do the following words describe the way you made your decision?

Deliberate Not at all 1 2 3 4 5 Strongly

Impulsive Not at all 1 2 3 4 5 Strongly

Spontaneous Not at all 1 2 3 4 5 Strongly

Well

thought-through Not at all 1 2 3 4 5 Strongly

To what extent did the following play a role in your decision?

Taking action Not at all 1 2 3 4 5 Strongly

Being careful Not at all 1 2 3 4 5 Strongly

Avoid mistakes Not at all 1 2 3 4 5 Strongly

Choosing an X that

was good enough Not at all 1 2 3 4 5 Strongly