Post on 22-Jul-2020
transcript
The Consequentialist Scale: Elucidating the Role of Deontological and Utilitarian Beliefs in Moral Judgments
by
Jeffrey Sean Robinson
A thesis submitted in conformity with the requirements for the degree of Master of Arts
Department of Psychology University of Toronto
© Copyright by Jeffrey Sean Robinson 2012
ii
The Consequentialist Scale: Elucidating the Role of Deontological and Utilitarian Beliefs in Moral Judgments
Jeffrey Sean Robinson
Master of Arts
Department of Psychology University of Toronto
2012 Abstract
In previous research, measurement of deontological and utilitarian beliefs relied on responses to
classic moral dilemmas. While use of these dilemmas has proven fruitful they are fraught with
potential confounds. This thesis describes the construction and validation of the Consequentialist
scale, a tool designed to directly measure the endorsement of both utilitarian and deontological
beliefs. The Consequentialist scale was tested against variables previously associated with moral
judgments, namely emotional reactivity, Machiavellianism, intrinsic religiosity, and political
conservatism. Results indicate that the Consequentialist scale provides added explanatory power
that helps to elucidate cornerstone processes involved in moral judgments.
iii
Acknowledgments
I would like to take this opportunity to thank both Dr. Jason Plaks and Dr. Elizabeth Page-Gould
for their guidance and support with this project. Thank you to Dr. Charles Helwig for sitting on
my thesis committee. I would also like to thank Dr. Adam Anderson for the opportunities he
afforded me which gave me the confidence to pursue my goals. Finally, a special thanks to my
parents, Steven and Cynthia Robinson, whose unconditional love and support have kept my
dreams alive.
iv
Table of Contents
Acknowledgments .......................................................................................................................... iii
Table of Contents ........................................................................................................................... iv
List of Tables ................................................................................................................................. vi
List of Figures ............................................................................................................................... vii
List of Appendices ....................................................................................................................... viii
Chapter 1 ......................................................................................................................................... 1
1 From Philosophy to Psychology ................................................................................................ 1
Chapter 2 ......................................................................................................................................... 5
2 Classifying Utilitarians and Deontologists: The Consequentialist Scale. ................................. 5
2.1 Problems of Measurement .................................................................................................. 5
2.2 Consequentialist Scale: Exploratory and Confirmatory Factor Analysis. ...................... 6
2.2.1 Method .................................................................................................................... 6
2.2.2 Results ..................................................................................................................... 8
2.3 Second Confirmatory Factor Analysis .............................................................................. 10
2.3.1 Method .................................................................................................................. 10
2.3.2 Results ................................................................................................................... 10
2.4 Discussion ......................................................................................................................... 11
Chapter 3 ....................................................................................................................................... 13
3 Reactivity, Cognitive Control and the Consequentialist scale: Rationale ............................... 13
3.1 Method: Study 1 ............................................................................................................... 14
3.2 Results ............................................................................................................................... 15
3.3 Method: Study 2 ............................................................................................................... 16
3.4 Results ............................................................................................................................... 17
3.5 Discussion ......................................................................................................................... 18
v
Chapter 4 ....................................................................................................................................... 20
4 Machiavellianism, Psychopathy, Life Meaninglessness and the Consequentialist scale: Rationale .................................................................................................................................. 20
4.1 Method .............................................................................................................................. 20
4.2 Results ............................................................................................................................... 21
4.3 Discussion ......................................................................................................................... 22
Chapter 5 ....................................................................................................................................... 24
5 Politics, Religion and the Consequentialist scale: Rationale .................................................. 24
5.1 Method .............................................................................................................................. 24
5.2 Results ............................................................................................................................... 26
5.3 Discussion ......................................................................................................................... 27
Chapter 6 ....................................................................................................................................... 29
6 Conclusions .............................................................................................................................. 29
References ..................................................................................................................................... 31
Appendices .................................................................................................................................... 35
vi
List of Tables
Table 1. Greene and Colleagues Dilemma Summary…………………………………………..35
Table 2. Unrefined Factor Loading Summary 1 – EFA………………………………………...37
Table 3. Refined Factor Loading Summary – EFA…………………………………………….40
Table 4. Moore and Colleagues Dilemma Summary…………………………………………...46
Table 5. Model Summary Section 3.4…………………………………………………………..51
Table 6. Correlation Summary Section 3.4……………………………………………………..52
vii
List of Figures
Figure 1. Scree Plot – EFA………………………………………………………………………39
Figure 2. Preliminary CFA Model 1……………………………………………………………..41
Figure 3. Preliminary CFA Model 2……………………………………………………………..42
Figure 4a. Predicting Wrongness Ratings: Two-way interaction………………………………..43
Figure 4b. Predicting Wrongness Ratings: Two-way interaction………………………………..43
Figure 5. Second CFA Model 1……………………………………………………………….....44
Figure 6. Second CFA Model 2……………………………………………………………….....45
Figure 7. Cognitive reflection, Deontological Beliefs and Wrongness Ratings: Mediation
Analysis………………………………………………………………………………………….48
Figure 8. Affect of Religion, Politics, and Consequentialist Beliefs: Model 1………………....49
Figure 9. Affect of Religion, Politics, and Consequentialist Beliefs: Model 2………………....50
viii
List of Appendices
Appendix A: Table 1 – Greene and Colleagues Dilemma Summary…………………………...35
Appendix B: Table 2 - Unrefined Factor Loading Summary 1 – EFA……………………….....37
Appendix C: Figure 1 – Scree Plot – EFA………………………………………………………39
Appendix D: Table 3. Refined Factor Loading Summary – EFA……………………………….40
Appendix E: Figure 2. Preliminary CFA Model 1…………………………………………….....41
Appendix F: Figure 3. Preliminary CFA Model 2…………………………………………….....42
Appendix G: Figure 4a & 4b. Predicting Wrongness Ratings: Two-way interaction…………...43
Appendix H: Figure 5. Second CFA Model 1…………………………………………………...44
Appendix I: Figure 6. Second CFA Model 2………………………………………………...…..45
Appendix J: Table 4 – Moore and Colleagues Dilemma Summary……………………………..46
Appendix K: Figure 7. Cognitive reflection, Deontological Beliefs and Wrongness Ratings:
Mediation Analysis……………..……………………………………………………………......48
Appendix L: Figure 8. Affect of Religion, Politics, and Consequentialist Beliefs: Model 1…...49
Appendix M: Figure 9. Affect of Religion, Politics, and Consequentialist Beliefs: Model 2…..50
Appendix N: Table 5. Model Summary Section 3.4……………………………………………..51
Appendix O: Table 6. Correlation Summary Section 3.4………………………………………..52
1
Chapter 1
1 From Philosophy to Psychology Interest in moral thought and reasoning has experienced a renaissance in recent times. In
particular, there as been increased interest in the psychology of utilitarianism, a topic that traditionally
had been studied primarily in philosophy and economics. The philosophical position of utilitarianism
popularized by Mill (1861) states:
Actions are right only insofar as they tend to produce the greatest balance of pleasure over pain for
the largest number of people.
From this position the only morally relevant feature of an action is whether or not it leads to a net
amount of good. Any action, even killing, is the morally correct thing to do if it leads to a net benefit.
The contrasting moral position is that of deontology, made famous by Kant (1785 –
Foundation of the metaphysics of morals). This philosophical position states:
Act only on the maxim through which you can at the same time will that it should become universal
law.
From this point of view, some actions are always wrong no matter the amount of good they may
produce. Killing is wrong because if made universally acceptable it would lead to a society defined by
a continual state of chaos.
Greene and colleagues (2001, 2004, 2008a, 2008b) brought the philosophical argument into
the psychological realm by asking what psychological mechanisms underlie the propensity to reason
in a utilitarian fashion. Greene, Sommerville, Nystrom, Darley, & Cohen (2001) were fascinated by
the apparent incongruity of people’s responses when asked to make judgments about utilitarian
dilemmas. The classic utilitarian dilemma asks participants whether it is morally permissible to divert
a runaway trolley car from a track that has five workmen on it to a track with only one workman on it,
thereby killing one but saving five (Foot, 1967). When presented with this dilemma, a large majority
of individuals say that it is morally acceptable to divert the trolley. However, when the trolley
dilemma is modified such that the only way to stop the runaway trolley from killing the five workmen
2
is to push a very large, fat, man onto the tracks most people say that this is not acceptable (Thomson,
1985). Strictly speaking, in both scenarios, whether you redirect the trolley or push the fat man, the
end is the same: one dies and five are saved. Yet people generally perceive important distinctions
between the two scenarios. Why is this? What are the psychological and neural mechanisms involved
in utilitarian versus deontological reasoning?
Processes Associated with Utilitarian and Deontological Thought
Greene, Nystrom, Engell, Darley, & Cohen (2004) used functional magnetic resonance
imaging (fMRI) and found that deontological choices were associated with stronger activation of areas
associated with emotional experience, namely the ventral medial prefrontal cortex (VMPFC) and
amygdala, relative to areas associated with deliberative reasoning such as the dorsal lateral prefrontal
cortex (DLPFC) and the anterior cingulate cortex (ACC). These findings support the view that a
propensity to reason in a deontological manner is associated with emotional reactivity. Furthermore,
Greene, Morelli, Lowenberg, Nystrom, & Cohen (2008a) provided evidence that utilitarian decisions
are under the control of deliberative reasoning processes. Participants responded to utilitarian
dilemmas under cognitive load. The results showed that individuals who made utilitarian decisions
(i.e. kill one to save five) took significantly longer to respond relative to those who made
deontological judgments (it is wrong to kill one person to save five) while under cognitive load. There
was no difference in response times between those who made utilitarian or deontological judgments in
the no cognitive load condition. As the cognitive load condition selectively affected reaction times of
only utilitarian choices, the data suggest that utilitarian judgments are more strongly associated with
deliberative reasoning and executive control.
Furthermore, as noted previously, utilitarian responses are associated with greater ACC
activation. Interestingly, the ACC is associated with the reconciliation of conflicting information
(Botvinick, Braver, Barch, Carter & Cohen, 2001). Deontological moral judgments requires an
individual to recognize that a rule applies in a given situation and to apply the rule accordingly;
therefore, systems involved in conflict resolution need not be engaged. Utilitarian judgments, on the
other hand, require individuals to evaluate whether causing some form of harm is justified in order to
promote some greater good. Therefore, individuals are forced to reconcile two conflicting
perspectives; on the one hand the idea that in general harm to others should be minimized and on the
other hand what is good for everyone should be maximized. Activation of the ACC during utilitarian
judgments indicates that these two conflicting principles are activated and their differential force is
3
weighed and reasoned out before making a moral judgment. Subsequently, Koenigs, Young,
Adolphs, Tranel, Cushman, and Hauser (2007) showed that individuals with damage to an area of the
brain associated with emotional experience (VMPFC) but no damage to regions associated with
deliberative reasoning make more utilitarian judgments. This evidence implies that specific cognitive
mechanisms that may underlie utilitarian responding.
Based on the evidence presented above it seems plausible that both trait emotional reactivity
and cognitive conflict regulation are important variables to consider in order to understand moral
judgments. These variables will be considered further in the analyses to come.
Traits Associated with Utilitarian and Deontological Thought
At a broader level of analysis, some researchers have used utilitarian moral judgments as the
normative standard (Sunstein, 2005; Baron & Ritov, 2009). In contrast to this position, Bartels and
Pizarro (2011) provided evidence that one’s propensity for utilitarian thinking is associated with
higher levels of psychopathy (characterized by low empathy, callous affect, and thrill-seeking),
Machiavellianism (the degree to which people are cynical, emotionally detached from others, and
manipulative), as well as ratings of life meaninglessness (how meaningful individuals perceive life to
be). This evidence suggests that utilitarians are comparatively egocentric, emotionally cold and
calculating individuals who perceive life as having little inherent meaning. This finding suggests a
disconnect between utilitarianism as a moral system and utilitarianism in practice, which is associated
with traits generally considered to be immoral.
I suggest that the trait approach adopted by Bartels and Pizarro cannot explain many types of
moral judgment. For example imagine a wartime scenario in which you and five of you allies are
pinned down by enemy fire in a bunker. A grenade is lobbed into the bunker and only you realize
what has happened. Several of your comrades are located between you and the grenade, and their
bodies will absorb the brunt of the explosion, so if you do nothing you will likely survive the blast but
many of your brothers in arms will be killed. However, you have just enough time to throw yourself
on the grenade, absorbing the explosion yourself, thereby saving your friends. A true utilitarian will
be forced to admit that the morally correct thing to do is to throw yourself on the grenade, despite this
option not being in your own best interest. It seems highly unlikely, however, that someone scoring
high on Machiavellianism would endorse making this kind of personal sacrifice. Based on this line of
reasoning, I suggest that – at the psychological level – two distinct types of utilitarian judgments may
exist, a prosocial type and an anti-social type. The methodology currently employed to assess
4
utilitarian beliefs is not designed to detect this subtle distinction.
Current Project
Based on the research reviewed here, it appears that there are several individual difference
variables that predict deontological and utilitarian moral judgments. However, the field has obtained
only a meager understanding of the relationship between emotional processes and the belief structure
that underlies these types of moral judgments and beliefs. For instance, it is commonly thought that
deontological and utilitarian beliefs represent the polar ends of a single dimension. This assumption,
however, has never been conclusively demonstrated.
This thesis sets out to elucidate a more comprehensive understanding of utilitarian and
deontological thinkers and the processes that lead to differential moral judgments. First, I will discuss
the creation and validation of the Consequentialist scale; a ten-item scale that measures an individuals
endorsement of both deontological and utilitarian values. Next, I will explore how the trait variables
discussed above interact with the Consequentialist scale to predict moral judgments. Lastly, I will
explore how the Consequentialist scale interacts with constructs regularly associated with moral
judgment, namely conservatism and religiosity, and the differential role these variables play in
influencing moral judgments.
5
Chapter 2
2 Classifying Utilitarians and Deontologists: The Consequentialist Scale. The first step was the development of a simple and easy to administer tool that would allow
research participants to be classified as having a propensity towards utilitarian or deontological
thought.
2.1 Problems of Measurement As far as this researcher is aware, past research examining utilitarian and deontological
thinking has required participants to evaluate classic utilitarian vignettes (e.g. Greene et al., 2001,
2008a, 2009; Koenigs et al., 2007; Rest, Narvaez, Thoma & Bebeau, 1999; Lind, 1978). This
approach is vulnerable to the criticism of circularity in that it uses dilemmas designed to illustrate a
moral principle to test for the presence or absence of the moral principle. Furthermore, while the
judgments made in response to these dilemmas may indicate a propensity for utilitarian (or
deontological) thought it provides no insight into the actual set of beliefs and values that led these
judgments. In addition some of the tools previously used involved moral judgments in highly specific
domains such as business (Casali, 2011; Zgheib, 2005; Reidenbach & Robin, 1988) and sports (Rudd,
Mullane, & Stoll, 2010). Lastly, the types of utilitarian dilemmas debated in moral philosophical
circles often involve highly elaborate scenarios that no actual human being has ever experienced and
only a select few have ever thought about (e.g., the trolley problem). These scenarios therefore lack
ecological validity, leaving the question of whether the pattern of responses observed are truly
reflective of the pattern of beliefs of the individual forced to make the choice, or if the responses are
simply an artifact of the artificiality of the situations proposed. While there is no doubt value in
seeking views on different types of moral situations, this approach suffers issues of control and
validity.
Furthermore, the types of vignettes used by researchers are highly variable across experiments.
Some authors have used highly emotionally charged items including taboos such as incest, eating
one’s dead dog (Haidt, 2001; Wheatley & Haidt, 2005) and cannibalism (Schnall, Haidt, Clore, &
Jordan, 2008) while others have used dilemmas that vary in terms of amount of personal force used
6
(Greene et al., 2009). Still others have varied whether the deaths of the victim(s) were avoidable or
inevitable, whether the actor personally benefited from their actions, and whether the actor acted
directly against the victim or used an object to act against the victim (i.e. pull a switch) (Moore, Clark,
& Kane, 2008).
With all this variability it is impossible to know for certain if peoples’ responses to these moral
dilemmas are a result of a latent propensity to utilitarian or deontological thought or are due to another
factor not controlled for across these studies. With these issues in mind, I sought to design a measure
that would assess utilitarian and deontological beliefs generally, without appealing to a specific
situation. The Consequentialist scale was designed to measure the extent to which individuals
endorse deontological (i.e. some rules should never be broken, regardless of the consequences that
result) and utilitarian (i.e. the only morally relevant dimension to judge an action is whether or not it
resulted in a net positive outcome) beliefs. The unrefined scale consisted of 40 statements in total, 20
that were designed to capture deontological beliefs and the remaining 20 designed to capture
utilitarian beliefs. Participants were told that they were going to read several different statements and
indicate to what level they agreed with each of the statements. Participants were asked to indicate
responses using a 5-point Likert scale (where 1 = Completely Disagree and 5 = Completely Agree).
This scale will provide a means other than the use of utilitarian dilemmas to assess consequentialist
beliefs while simultaneiously improving our understanding of deontological and utilitarian moral
judgments by providing insight into the specific belief structures involved in making moral judgments.
In subsequent sections I describe the validation procedures used to construct the final version of the
Consequentialist scale.
2.2 Consequentialist Scale: Exploratory and Confirmatory Factor Analysis.
2.2.1 Method Participants. A community sample of 488 residents of the United States of America was
collected using Mechanical Turk (operated by Amazon.com). Participants were compensated $0.50
for their participation. The sample consisted of 270 females (55.3%) and 215 males (44.1%). Three
individuals identified their gender as being “other” with two individuals explicitly stating that they
identify as transgendered. The average age of participants was 30.72 years (SD = 11.14). Twenty-five
percent of the sample was over the age of 37.17 years and 50% of the sample was under 27.3 years of
age. Participants reported an average of 15.16 years of formal education starting from grade one (SD
7
= 3.24). Lastly, the median income range for participants was between $10,000 and $25,000. In
terms of income distribution, 57.53% of participants reported earning $25,000 or less in the last
calendar year while 16.6% reported earning $50,000 or more in the last calendar year.
Procedure. Participants logged on to Mechanical Turk and were randomly assigned to one of
three different surveys. Each survey was designed to assess how deontological and utilitarian beliefs
interact with variables previously associated with moral judgments namely, 1) emotional reactivity, 2)
Machiavellianism, Psychopathy, Life Meaninglessness, and 3) Political ideology and religiosity.
Within each survey participants completed a demographics form, the Consequentialist scale, and read
and responded to six different utilitarian dilemmas. Responses to the demographics form, the
Consequentialist scale and the utilitarian dilemmas were merged to form one large data set in order to
confirm the factor structure of the Consequentialist scale.
Consequentialist scale. This scale was designed to measure the extent to which individuals
endorse deontological and utilitarian beliefs. As consequentialism has never been analyzed in this
fashion, I performed a multistage process to develop a new Consequentialist Beliefs Scale. First, I
hypothesized that consequentialist beliefs would be best understood as consisting of two factors: 1)
Deontological beliefs (which emphasize that some rules are universal and should never be broken) and
2) Utilitarian beliefs (which stress the only relevant factor when deciding whether an action is morally
correct is whether it leads to a net benefit). From this hypothesis 40 items were deductively generated,
with 20 items attempting to tap into deontological thought and 20 items attempting to tap utilitarian
ideology. All 40 items were combined into an online survey using Survey Monkey. Participants were
told that they were going to read several different statements and indicate to what level they agreed
with each of the statements. Participants were asked to indicate responses using a 5-point Likert scale
(where 1 = Completely Disagree and 5 = Completely Agree).
Dilemmas. Classic moral dilemmas often involve the extreme act of killing which most
participants would rate as severely wrong. I was wary of using scenarios that most people would rate
as severely wrong and therefore to prevent ceiling effects for wrongness ratings I wanted to select
vignettes that were highly contentious in the sense that there would be great disagreement as to
whether killing was justified in the story depicted. I selected a sub-set of high conflict dilemmas used
by Greene, Morelli, Lowenberg, Nystrom, and Cohen (2008a). Using the results from Greene and
colleagues (2008) six dilemmas were identified where participants made the utilitarian judgment (i.e.
it is appropriate to sacrifice one to save many) on average 57% of the time (See Table 1. Appendix
A.). Given that judgments of the appropriateness of the action were split I felt that these dilemmas
8
were sufficiently contentious to prevent ceiling effects. After reading each of the vignettes participants
were asked, “Based on what you just read, how wrong would it be for you [to act in a utilitarian
fashion, using the specifics of the scenario they had just read]?” Participants were asked to indicate
their responses using a 7-point Likert scale (where 1 = not wrong at all and 7 = completely wrong).
2.2.2 Results Exploratory factor analysis.
To test the 2-factor structure of utilitarian and deontological beliefs, I conducted an exploratory
factor analysis using the maximum likelihood extraction method, indicating two factors and utilized a
promax rotation. The promax rotation was selected as I expected the two factors to be correlated. The
initial factor analysis and examination of the pattern matrix revealed that several items did not load
significantly on either of the two factors (for all values see Table 2. – Appendix B), and as a result any
item that did not load greater than 0.4 was eliminated. The analysis was run again using only the
items that had loaded significantly in the previous analysis and a stable two-factor structure emerged.
Examination of the scree plot (Figure 1 – Appendix C) revealed a steep increase in eigenvalues with
only two factors remaining. The eigenvalue for the factor “Utilitarian Beliefs” was 3.718 and the
eigenvalue for the factor “Deontological Beliefs” was 2.769. Based on these values a two-factor
solution was justified.
The five highest loading items for each factor were identified and retained for confirmatory
factor analysis (see Table 3 – Appendix D). None of the items had significant cross-loadings between
factors. The five items that loaded most strongly onto “Deontological Beliefs” were shown to be
reliable, α = 0.69, and the five-items that most strongly loaded onto “Utilitarian Beliefs” were shown
to be highly reliable, α = 0.8.
Confirmatory factor analysis. To provide additional support for the 2-factor structure of the
Consequentialist scale a confirmatory factor analysis was conducted using SPSS AMOS 20. A model
(see Figure 2 – Appendix E) was generated with two latent variables (Deontological and Utilitarian
Beliefs). Tests of fit revealed that the model had good fit, RMSEA = 0.066 90% CI [0.05, 0.08], CFI
= 0.93, AIC = 168.18. An absolute goodness of fit test was inappropriate in this case as the sample
size was well above 400 participants (Stevens, 1996). These results support the validity of the 2-factor
structure of the Consequentialist scale.
Based on the literature I hypothesized that deontological and utilitarian beliefs would be
significantly, negatively correlated with each other. While this assumption is ubiquitous to this
9
researcher’s knowledge it has never been tested explicitly. To test this assumption a second model was
run in which the covariance between the latent constructs of deontological and utilitarian beliefs was
set to zero. Tests of fit for this second model (See Figure 3 – Appendix F) revealed good fit, RMSEA
= 0.65 90% CI [0.05, 0.079], CFI = 0.93, AIC 167.74. The nested model comparison test was not
significant, χ2(1) = 1.568, p = 0.21, indicating both models fit the data equally well. For the sake of
parsimony the second model was selected for future use. These results indicate that the assumption
that deontological and utilitarian beliefs are negatively correlated may not be accurate.
Predicting wrongness judgments. In the final stage of this analysis I examined whether the
two sub-scales of the Consequentialist Scale would predict participants’ moral judgments of the
utilitarian dilemmas. In other words, greater endorsement of deontological beliefs should predict
higher wrongness ratings and greater endorsement of utilitarian beliefs should predict lower
wrongness ratings. Ratings of “wrongness” across the six dilemmas were highly reliable, α = 0.793.
Therefore an aggregate wrongness rating score was calculated. Given that the confirmatory factor
analysis indicated that deontological and utilitarian beliefs were not significantly correlated, a model
was tested where wrongness ratings were regressed on deontological beliefs, utilitarian beliefs, and
their interaction term. This model revealed a significant main effect of deontological beliefs, b =
0.624, SE = 0.072, t(484) = 8.688, p<0.0001, η2Part
= 0.362, a significant main effect of utilitarian
beliefs, b = -0.223, SE = 0.072, t(484) = -3.109, p = 0.002, η2Part
= -0.129, and a significant
interaction, b = -0.115, SE = 0.048, t(484) = -2.4, p = 0.017, η2Part
= -0.1 (See Figure 4a – Appendix
G). Adding the interaction term to the model accounted for significantly more variance in comparison
to the model that omitted the interaction, R2change = 0.01, F(1, 484) = 5.762, p = 0.017. Analysis of the
simple effects revealed a significant effect of utilitarian beliefs when deontological beliefs were high,
b = -0.287, SE = 0.071, t(484) = -4.059, p < 0.0001, but not when deontological beliefs were low, b = -
0.057, SE = 0.076, t(484) = -0.75, p = 0.454. Furthermore, there was a significant effect of
deontological beliefs both when utilitarian beliefs were high, b = 0.364, SE = 0.073, t(484) = 4.984, p
< 0.0001, and when utilitarian beliefs were low, b = 0.594, SE = 0.073, t(484) = 8.122, p < 0.0001. In
summary greater endorsement of deontological beliefs led to increased wrongness ratings regardless
of the level of endorsement of utilitarian beliefs. In addition, wrongness ratings were most severe
when endorsement of utilitarian beliefs was low and endorsement of deontological beliefs was high.
10
2.3 Second Confirmatory Factor Analysis The exploratory and confirmatory analysis conducted above provides evidence for the validity
of the Consequentialist scale. In order to confirm the structure of the 10-item refined version of the
Consequentialist scale a second independent sample of data was collected and a second confirmatory
factor analysis was conducted.
2.3.1 Method Participants. A community sample of 658 residents of the United States of America was
collected using Mechanical Turk (operated by Amazon.com). Participants were compensated $0.50
for their participation. The sample consisted of 395 females (57.8%) and 261 males (38.2%). Two
individuals identified their gender as being “other” and 25 individuals failed to indicate their gender.
The average age of participants was 34.42 years (SD = 12.02). Twenty-five percent of the sample was
over the age of 38 years and 50% of the sample was under 27 years of age. Participants reported an
average of 14.97 years of formal education starting from grade one (SD = 2.735). Lastly, the median
income range for participants was between $10,000 and $25,000. In terms of income distribution,
55.8% of participants reported earning $25,000 or less in the last calendar year while 16.4% reported
earning $50,000 or more in the last calendar year.
Procedure. The procedure was simpler than the previous study. Participants logged on to
Mechanical Turk and consented to participate in a study examining different thinking styles. Within
the survey participants completed a demographics form and the Consequentialist scale, and also read
and responded to six utilitarian dilemmas. These surveys completed were identical to those used in
section 2.2.
2.3.2 Results Confirmatory factor analysis. A confirmatory factor analysis (CFA) was conducted using
AMOS. A model (see Figure 5 – Appendix H) was generated with two latent variables (Deontological
and Utilitarian Beliefs). Tests of fit revealed that the model had good fit, RMSEA = 0.057 90% CI
[0.045, 0.069], CFI = 0.955, AIC = 170.376. An absolute goodness of fit test was inappropriate in this
case as the sample size was well above 400 participants (Stevens, 1996). These results support the
validity of the 2-factor structure of the Consequentialist scale in a large independent sample.
While it was hypothesized that the latent constructs in this model would be correlated, results
from the first CFA hinted that endorsement of utilitarian and deontological beliefs may be orthogonal
11
dimensions. To explicitly test this hypothesis a second model was run in which the covariance
between deontological and utilitarian beliefs was set to zero. Tests of fit for this secondary model
(See Figure 6 – Appendix I) revealed good fit, RMSEA = 0.061 90% CI [0.049, 0.072], CFI = 0.947,
AIC 182.76. The nested model comparison test was highly significant, χ2(1) = 14.382, p < 0.0001,
indicating that the model in which utilitarian and deontological beliefs were assumed to correlate fit
the data significantly better then when it was assumed that they were uncorrelated. While at first
glance this may appear contradictory to previous results, this is not the case. The nested model
comparison test from the first CFA indicated that the models fit the data equally well; the uncorrelated
model was only selected due to parsimony. This second CFA seems to confirm the field’s prevailing
assumption that deontological and utilitarian beliefs are significantly correlated.
Predicting wrongness judgments. In the final stage of this analysis I examined whether the
two sub-scales of the Consequentialist Scale were predictive of participants’ moral judgments in
reaction to the exposure to utilitarian dilemmas. This analysis indicated that ratings of “wrongness”
across the six dilemmas were highly reliable, α = 0.801. Therefore I calculated an aggregate
wrongness rating score. Given that the first confirmatory factor analysis conducted indicated that
deontological and utilitarian beliefs were not significantly correlated, a model was tested where
wrongness ratings were regressed on deontological beliefs, utilitarian beliefs, and their interaction
term. The analysis revealed a significant main effect of deontological beliefs, b = 0.657, SE = 0.159,
t(516) = 4.141, p < 0.0001, η2Part
= -0.162, but no significant main effect of utilitarian beliefs, b = -
0.262, SE = 0.224, t(516) = -1.172, p = 0.242, η2Part
= -0.046, and no significant interaction, b = 0.014,
SE = 0.066, t(516) = 0.204, p = 0.838, η2Part
= 0.008 (See Figure 4b – Appendix G). The addition of
the interaction term did not account for significantly more variance than the model that only included
main effects of deontological and utilitarian beliefs, R2change < 0.0001, F(1, 516) = 5.762, p = 0.838.
These findings indicate that deontological beliefs best predict wrongness ratings and there is no
significant interaction between deontological and utilitarian beliefs.
2.4 Discussion
This analysis provides evidence that the 10-item Consequentialist scale is a valid measure of
endorsement of both deontological and utilitarian beliefs. The data from the first CFA revealed a
significant main effect of both deontological and utilitarian beliefs as well as a significant interaction
when prediciting wrongness ratings. The data from the second CFA revealed a significant main effect
of deontological beliefs when prediciting wrongness ratings. This demonstrates that the
12
Consequentialist scale provides explanatory power in regards to moral judgments. The
Consequentialist scale provides a tool to assess consequentialist beliefs while avoiding the issues and
confounds associated with the use of classic moral dilemmas.
One issue that is yet to be resolved is whether deontological and utilitarian beliefs are best
understood as independent or related constructs. The first CFA indicated that deontological and
utilitarian beliefs were uncorrelated and in fact significantly interacted. However, the second CFA
favoured a model in which the deontological and utilitarian beliefs were significantly associated.
More data is required to make a clear assertion as to the relationship between deontological and
utilitarian beliefs.
13
Chapter 3
3 Reactivity, Cognitive Control and the Consequentialist scale: Rationale Haidt’s (2001) Social Intuitionist model asserts that moral judgments are the product of
intuitive, unconscious and automatic emotional responses. Research has demonstrated that moral
judgments are linked with emotional experience. Chapman, Kim, Susskind, and Anderson (2009)
found that the facial response to a moral transgression in the fairness domain was similar to that of an
individual tasting a bitter substance, thus linking fairness judgments with the experience of disgust.
Cannon, Schnall and White (2011) found that wrongness ratings of transgressions that involved harm
were predicted most accurately by activation of the corrugator muscles of the face, linking these
judgments to the experience of anger, while transgressions in the fairness and purity domains were
best predicted by activation of the levator labii, linking these judgments to the experience of disgust.
Furthermore, Greene and colleagues (2001, 2004, 2008) have shown that utilitarian decision-making
is associated with reduced activation of brain regions associated with emotional experience and
increased activation of regions associated with cognitive control. These findings all support the notion
that moral reasoning is a function of one’s ability to detect and regulate emotional information, which
in turn is a function of an individual’s sensitivity to stimuli from both bodily sensation and external
stimulation. In study 1 of this chapter I will explore the relationship between emotional reactivity, the
Consequentialist scale, and moral judgments. Based on this research it is hypothesized that
endorsement of utilitarian beliefs will be negatively related to emotional reactivity while endorsement
of deontological beliefs will be positively correlated with emotional reactivity.
In a different vein, Greene and colleagues (2001, 2004, 2008) propose a dual-system approach
to utilitarian thinking in which utilitarian decisions are the result of cognitive control mechanisms that
are thought to suppress emotional systems. As a measure of cognitive control Moore, Clark, and
Kane (2008) show that increased utilitarian responding is associated with greater working memory
capacity. Furthermore, Bartels (2008) demonstrated that individuals with more “rational” thinking
styles were more likely to make utilitarian moral judgments. Evidence also suggests that when people
take time to reflect on the moral scenarios they are presented with, judgments are affected. Paxton,
Ungar, and Greene (2011) showed that argument strength influenced moral judgments, but only when
14
participants were given time to reflect on the arguments. Suter and Hertwig (2011) showed that when
cognitive control was inhibited, by making participants respond quickly, participants made more
deontological decisions. In study 2 of this chapter I will explore the relationship between cognitive
control, the Consequentialist scale and moral judgments. Based on this research it is hypothesized
that endorsement of utilitarian beliefs will be positively be correlated with measures of cognitive
control while there will be no relationship with endorsement of deontological beliefs.
3.1 Method: Study 1 Participants. A community sample of 165 residents of the United States of America was
collected using Mechanical Turk (operated by Amazon.com). Participants were compensated $0.50
for their participation. The sample consisted of 90 females (55.3%) and 74 males (44.8%). One
individual (0.6%) identified their gender as being “other”. The average age of participants was 30.37
years (SD = 11.123). Twenty-five percent of the sample was over the age of 35.5 years and 50% of
the sample was under 26 years of age. Participants reported an average of 15.23 years of formal
education starting from grade one (SD = 3.201). Lastly, the median income range for participants was
between $10,000 and $25,000. In terms of income distribution, 59.4% of participants reported earning
$25,000 or less in the last calendar year while 17.5% reported earning $50,000 or more in the last
calendar year.
Procedure. Participants logged on to Mechanical Turk and were randomly assigned to one of
three different surveys 1) reactivity, 2) social interaction and 3) higher order beliefs. The data to be
discussed here is from the reactivity survey. Within each survey participants completed a
demographics form and the Consequentialist scale, and also read and responded to the same six
utilitarian dilemmas used in the exploratory and confirmatory factor analysis for the Consequentialist
Scale in chapter 2. In order to test our specific hypothesis, participants also completed the Three-
Domain Disgust Scale (Tybur, Griskevicius & Lieberman, 2009).
Three-Domain Disgust scale. Tybur and colleagues (2009) purpose that disgust sensitivity
can be sub-divided into three functional domains related to 1) the avoidance of harmful pathogens
(pathogen disgust) 2) the avoidance of sexual contact with biologically costly partners (sexual disgust)
and 3) avoiding individuals that may harm us or our kin (moral disgust). In this study I was interested
in the role of basic emotional reactivity and therefore scores on the pathogen disgust sub-scale were
optimal for this analysis. Furthermore, due to the morally laden nature of the sexual and moral
domains of the disgust scale they were omitted from the analysis. The pathogen disgust sub-scale
15
consisted of seven items that were rated on a seven-point Likert scale (1 = Not disgusting at all, 7 =
Extremely disgusting). Analysis revealed that the seven items were highly reliable, α = 0.86.
3.2 Results In order to further validate the Consequentialist scale I first inspected the relationship between
wrongness ratings and the endorsement of both deontological and utilitarian beliefs. First,
deontological and utilitarian beliefs were not significantly correlated, r(163) = -0.028, p = 0.725.
Subsequently, wrongness ratings across the six vignettes were shown to be highly reliable, α = 0.832,
therefore an aggregate “wrongness rating” variable was created. Next, wrongness ratings were
regressed on deontological beliefs, utilitarian beliefs and their interaction the analysis revealed a
significant main effect of deontological beliefs, b = 0.775, SE = 0.129, t(161) = 6.023, p < 0.0001,
η2Part
= 0.424, but no main effect of utilitarian beliefs, b = -0.183, SE = 0.138, t(161) = -1.327, p =
0.187, η2Part
= -0.093, and no significant interaction, b = -0.183, SE = 0.143, t(161) = -1.275, p = 0.204,
η2Part
= -0.09. The addition of the interaction term failed to account for significantly more variance
than the model that included only main effect of deontological and utilitarian beliefs, R2change = 0.008,
F(1, 207) = 1.626, p = 0.2. Furthermore, deontological and utilitarian beliefs were uncorrelated, r(163)
= -0.028, p = 0.725. These results indicate that deontological and utilitarian beliefs are unrelated and
deontological beliefs are the best predictor of wrongness ratings.
Next I tested whether pathogen disgust was related to wrongness ratings and whether the
Consequentialist scale remained a significant predictor when pathogen disgust was added to the
model. To ensure the Consequentialist scale was accounting for unique variance wrongness ratings
were regressed simultaneously on pathogen disgust and both sub-scales of the Consequentialist Scale.
This test revealed a significant main effect of pathogen disgust, b = 0.288, SE = 0.082, t(161) = 3.524,
p = 0.001, η2Part
= 0.24, endorsement of deontological beliefs, b = 0.719, SE = 0.126, t(161) = 5.715, p
< 0.0001, η2Part
= 0.389, and a significant main effect of endorsement of utilitarian beliefs, b = -0.263,
SE = 0.133, t(161) = -1.975, p = 0.05, η2Part
= -0.134.
These results indicate that the three predictors accounted for significant but independent
variance in wrongness ratings. While a mediating relationship was hypothesized, the variables did not
meet the criteria allowing for a mediation analysis. Namely, pathogen disgust was not significantly
related to either deontological, r(163) = 0.133, p =0.08, or utilitarian beliefs, r(163) = 0.11, p = 0.16.
These results provide evidence that contrary to Haidt’s social intuitionist model moral beliefs (in this
16
case endorsement of deontological and/or utilitarian values) provide vital explanatory insight, over and
above emotional reactivity, in regards to how individuals make moral judgments.
3.3 Method: Study 2
The previous study demonstrated the relationship between emotional reactivity and and the
two sub-scales of the Consequentialist scale. As discussed above the role of cognitive control has also
been implicated as vital to understanding how people reconcile moral dilemmas. In the current study I
explore the relationship between cognitive control and the Consequentialist scale and how this
relationship influences moral judgments. The finding of Greene and collegues (2001, 2004, 2008)
suggest that greater cognitive control will be associated with greater endorsment of utilitarian beliefs
and less endorsment of deontological beliefs.
A secondary goal of the current study was to demonstrate the utility of the Consequentialist
scale across different moral dilemmas. With that goal in mind participants responed to a different set
of utilitarian dilemmas.
Participants. A community sample of 212 residents of the United States of America was
collected using Mechanical Turk (operated by Amazon.com). Participants were compensated $0.50
for their participation. The sample consisted of 148 females (63.8%) and 84 males (36.2%). The
average age of participants was 31.55 years (SD = 11.937). Twenty-five percent of the sample was
over the age of 39 years and 50% of the sample was under 27.5 years of age. 37.1% of participants
reported having earned a high school diploma or less while 62.9% of respondents had earned a
Bachelor’s degree (or college equivalent) or higher. Lastly, the median income range for participants
was between $10,000 and $25,000. In terms of income distribution, 50.9% of participants reported
earning $25,000 or less in the last calendar year while 20.2% reported earning $50,000 or more in the
last calendar year.
Procedure. Participants logged on to Mechanical Turk and consented to participate in a study
about different styles of thinking. Within the survey participants completed the Cognitive Reflection
Task, the Consequentialist Scale, six moral dilemmas, and a demographics form.
Dilemmas. In order to strengthen the development of the Consequentialist scale as a valid
measure of moral beliefs a new set of six moral dilemmas was chosen for this study. The dilemmas
used were originally published by Moore and colleagues (2008). In order to prevent ceiling effects for
wrongness ratings of the moral dilemmas we wanted to select vignettes that were highly contentious in
17
the sense that there was high disagreement as to whether the individual depicted in the story was
justified in killing. Pilot testing was conducted asking participants to evaluate all of the original
dilemmas used by Moore and colleagues (2008). The dilemmas that were selected were personal
dilemmas that had close to a 50/50 split in the ratings of appropriateness for the actions depicted. The
seven dilemmas that were selected were shown to be highly reliable, α = 0.925, (See Table 4 –
Appendix J) therefore an aggregate measure for wrongness ratings was calculated.
Cognitive Reflection Task. The Cognitive Reflection task (CRT) (Fredrick, 2005) consists of
three word problems that are designed to illicit an intuitively appealing but false response. In order to
answer the problem correctly participants must override the intuitively appealing response. Based on
the findings of Greene and colleagues utilitarian moral judgments involve an analogous process in
which the intuitive emotional response is overridden. Furthermore, Paxton, Ungar, and Greene (2011)
have shown that the scores on the cognitive reflection task are predictive of more utilitarian moral
judgments. Based on these results, I predicted that CRT scores should be positively correlated with
endorsement of utilitarian beliefs and negatively correlated with endorsement of deontological beliefs.
3.4 Results The first question was whether the Consequentialist scale would predict wrongness ratings for
the new set of dilemmas. Wrongness ratings were regressed on endorsement of deontological and
utilitarian beliefs as well as their interaction term. The results showed that both endorsement of
deontological beliefs, b = 0.829, SE = 0.144, t(207) = 5.766, p < 0.0001, η2Part = 0.367, and
endorsement of utilitarian beliefs, b = -0.282, SE = 0.136, t(207) = -2.075, p = 0.039, η2Part = -0.132,
were significant predictors of wrongness ratings. There was no significant interaction between
deontological and utilitarian beliefs, b = -0.006, SE = 0.186, t(207) = -0.031, p = 0.975, η2Part = -0.002
The additional interaction term did not account for significantly more variance, R2change < 0.0001, F(1,
207) = 5.762, p = 0.975. Furthermore, deontological and utilitarian beliefs were uncorrelated, r(232) =
-0.101, p = 0.123.
The next hypothesis tested whether higher scores on the CRT would be associated with more
utilitarian moral judgments. A simple bivariate correlation revealed that CRT scores was significantly
associated with lower wrongness ratings across vignettes, r(209) = -0.166, p = 0.016, and hence more
utilitarian moral judgments. In order to assess whether the Consequentialist scale would predict
wrongness ratings over and above the CRT wrongness ratings were regressed on a full factorial model
that included CRT score, endorsement of deontological beliefs, and endorsement of utilitarian beliefs
18
and all interaction terms. The results revealed a main effect of deontological beliefs, b = 0.786, SE =
0.154, t(203) = 5.1, p < 0.0.0001, η2Part = 0.325, and a main effect of utilitarian beliefs, b = -0.328, SE
= 0.146, t(203) = -.155, p = 0.026, η2Part = -0.143. The previously significant association between CRT
scores and wrongness ratings was no longer present in the full factorial model, b = -0.109, SE = 0.096,
t(203) = -1.129, p = 0.26, η2Part = -0.072. Possible interaction effects were tested for but no significant
two- or three-way interactions were seen (See Table 5 - Appendix N).
The analysis above hinted at a possible mediating relationship of Consequentialist beliefs
between CRT scores and wrongness ratings. Finally, because there were significant inter-correlations
between wrongness ratings, endorsement of deontological beliefs (but not utilitarian beliefs), and CRT
scores (See Table 6 – Appendix O) a mediation analysis was conducted by using bootstrapping to
estimate the size and standard errors of both direct and indirect pathways (Preacher & Hayes, 2008).
5000 bootstrap resamples were used to provide stable estimates of the direct, indirect, and total effects.
95% confidence intervals were determined from bootstrap resamples and any interval that did not
include zero was considered to be significantly different from zero. The analysis revealed that CRT
scores influenced wrongness ratings as a function of endorsement of deontological beliefs, ab = -
0.1554, SE = 0.044, 95% CI [-0.2552, -0.0818]. While the direct effect (not taking into account the
mediator) of CRT scores on wrongness ratings was significant, c = -0.2352, SE = 0.0966, t(210) = -
2.4347, p = 0.0157, this effect disappeared when endorsement of deontological beliefs was taken into
account, c’ = -0.0802, SE = 0.095, t(209) = -0.8446, p = 0.3993, indicating complete mediation (See
Figure 7 – Appendix K). In summary I replicated the findings of Paxton and colleagues, but the
mediation analysis revealed that the relationship between wrongness ratings and cognitive reflexivity
was due to higher CRT scores being associated with less endorsement of deontological beliefs. CRT
scores had no direct influence on endorsement of utilitarian beliefs as predicted.
3.5 Discussion
The results of these studies lend additional support for the functionality and validity of the
Consequentialist scale. In the first study deontological beliefs, but not utilitarian beliefs, were
significant predictors of wrongness ratings while in the second study wrongness ratings were
significantly predicted by both deontological and utilitarian beliefs. Furthermore, in both studies
deontological and utilitarian beliefs were not significantly associated with each other nor did they
interact to influence wrongness ratings. These results support the predicitive validity of the
19
Consequentialis scale and also supports the view that deontological and utilitarian beliefs represent
independent dimensions.
The results of study 1 replicate previous findings indicating that emotional reactivity is
associated with harsher moral judgments. However, these results add to our understanding by
showing that the relationship between emotional reactivity and moral judgments is not related to
deontological or utilitarian beliefs. While previous work has connected disgust sensitivity with higher
levels of rule-based thinking and conservative moral judgments (Helzer & Pizarro, 2011; Terrizzi,
Shook, & Ventis, 2010; Inbar, Pizarro, Knobe, & Bloom, 2009) the evidence presented here suggests
that these findings cannot be explained by appealing to endorsement of deontological beliefs. The
results suggest that emotional reactivity and deontological beliefs are independent contributors to
moral judgment severity.
In study 2 the goal was to elucidate the relationship between cognitive control and the
Consequentialist scale. Those who scored higher on the CRT were more likely to make utilitarian
moral judgments. The mediation analysis revealed that the effect of the CRT on wrongness ratings
was mediated by deontological beliefs, with greater CRT scores being associated with lower
endorsement of deontological beliefs. Counter to prediction, endorsement of utilitarian beliefs had no
significant relationship with CRT scores. This finding indicates that utilitarian moral judgments may
be related more to the absence of deontological beliefs rather than the presence of utilitarian beliefs.
Taken together these results illustrate a complex relationship between emotional reactivity,
cognitive control, and consequentialist beliefs, giving us a better understanding of how reactivity and
cognition contribute to moral judgments.
20
Chapter 4
4 Machiavellianism, Psychopathy, Life Meaninglessness and the Consequentialist scale: Rationale
Evidence has begun to emerge that patterns of moral judgment consistent with utilitarianism
are linked to personality traits such as Psychopathy, Machiavellianism, and ratings of Life
Meaninglessness, all of which are traditionally thought to be immoral (Bartels and Pizarro, 2011).
While these traits have been associated with evaluating moral dilemmas in a utilitarian fashion it
remains unclear whether these judgments are motivated by an appeal to morally principled utilitarian
beliefs or some other set of possibly immoral values. In this section, the relationship between
Consequentialist beliefs and Machiavellianism, Psychopathy, Life Meaninglessness and moral
judgments will be explored. I hypothesized that Machiavellianism, Psychopathy, and Life
Meaninglessness would all be associated with with more utilitarian moral judgments, and this
relationship would be mediated by endorsement of utilitarian beliefs and condemnation of
deontological beliefs. This mediation model would provide evidence that utilitarian beliefs provide
explanatory power over and above, Machiavellianism, psychopathy, and life meaninglessness when it
comes to moral judgments. This in turn would provide evidence that utilitarian beliefs are not
synonomous with Machiavellianism, psychopathy, and life meaninglessness. Although there is
overlap between utilitarianism and these constructs– especially in the sense that “the end justifies the
means” – utilitarianism is ultimately a moral system that is concerned with promoting the general
good, whereas Machiavellians and psychopaths, by definition, are primarily concerned with advancing
their own personal aims.
4.1 Method Participants. A community sample of 165 residents of the United States of America was
collected using Mechanical Turk (operated by Amazon.com). Participants were compensated $0.50
for their participation. The sample consisted of 89 females (53.9%) and 75 males (45.5%). One
individual (0.6%) identified their gender as being “other”. The average age of participants was 31.15
years (SD = 11.55). Twenty-five percent of the sample was over the age of 38 years and 50% of the
sample was under 27 years of age. Participants reported an average of 15.42 years of formal education
21
starting from grade one (SD = 3.269). Lastly, the median income range for participants was between
$10,000 and $25,000. In terms of income distribution, 58.8% of participants reported earning $25,000
or less in the last calendar year while 12.7% reported earning $50,000 or more in the last calendar
year.
Procedure. Participants logged on to Mechanical Turk and were randomly assigned to one of
three different surveys 1) reactivity, 2) social interaction and 3) higher order beliefs. The data to be
discussed here is from the social interaction survey. Within each survey participants completed a
demographics form, the Consequentialist scale, and read and responded to the same six utilitarian
dilemmas used in chapter two and section 3.1. Wrongness ratings for the dilemmas were shown to be
highly reliable, α = 0.78, and so an aggregate measure of wrongness ratings was created for use in the
analyses below. As for the Consequentialist scale, both the deontological belief sub-scale, α = 0.706,
and the utilitarian belief sub-scale, α = 0.82, were found to be reliable. In order to test our specific
hypothesis participants also completed the short Dark Triad questionnaire (Paulhus & Williams, 2002)
and the Life Meaninglessness scale (Kunzendorf, Moran, & Gray, 1996). The subscale of the Dark
Triad questionnaire, Machiavellianism, α = 0.763, and Psychopathy, α = 0.753, were found to be
reliable. Finally, the items of the Life Meaninglessness scale were also found to be reliable, α = 0.935.
4.2 Results First, we set out to provide additional support for the reliability and validity of the
Consequentialist scale. Regression analysis revealed a significant main effect of deontological beliefs,
b = 0.956, SE = 0.360, t(161) = 2.652, p = 0.009, η2Part = -0.193, but no significant main effect of
utilitarian beliefs, b = 0.275, SE = 0.451, t(161) = 0.61, p = 0.542, η2Part = 0.042, and no significant
interaction, b = -0.171, SE = 0.141, t(161) = -1.21, p = 0.228, η2Part = -0.088. Furthermore, the model
that included the interaction term did not account for significantly more variance when compared to
the model that included only the main effects of deontological and utilitarian beliefs, R2change = 0.008,
F(1, 161) = 1.465, p = 0.23. Furthermore, there was no significant association between deontological
and utilitarian beliefs, r(163) = -0.07, p = 0.37.
Next we set out to replicate the result of Bartels and Pizarro (2011). Consistent with Bartels
and Pizarro’s results, wrongness ratings were significantly correlated with Machiavellianism, r(163) =
-0.286, p < 0.0001, Psychopathy, r(163) = -0.182, p = 0.019, and Life Meaninglessness, r(163) = -
0.231, p = 0.003, showing individuals scoring high on these measures made less severe moral
judgments (i.e. it was less wrong to sacrifice one in order to save many).
22
Next, in order to assess the unique contribution of Machiavellianism, Psychopathy, and Life
Meaninglessness on wrongness ratings a regression analysis was conducted. Wrongness ratings were
regressed simultaneously on the three traits. All predictor variables were mean centered prior to that
analysis. Wrongness ratings were only significantly predicted by Machiavellianism, b = -0.528, SE =
0.21, t(161) = -2.520, p = 0.01, η2Part = -0.189, but not Psychopathy, b = 0.062, SE = 0.203, t(161) =
0.306, p = 0.76, η2Part = 0.023, or Life Meaninglessness, b = -0.297, SE = 0.175, t(156) = -1.69, p =
0.093, η2Part = -0.127. Machiavellianism was therefore the only significant predictor of wrongness
ratings when controlling for Psychopathy and Life Meaninglessness. For this reason Psychopathy and
Life Meaninglessness will not be discussed further.
The next question to be addressed was whether consequentialist beliefs are a significant
predictor over and above Machiavellianism. To test this hypothesis, wrongness ratings were regressed
simultaneously on Machiavellianism and endorsement of deontological beliefs and utilitarian beliefs.
All predictor variables were mean centered prior to the analysis. Both Machiavellianism, b = -0.465,
SE = 0.161, t(161) = -2.863, p = 0.005, η2Part = -0.204, and endorsement of deontological beliefs, b =
0.502, SE = 0.117, t(161) = 4.281, p < 0.0001, η2Part = 0.305, were significant predictors of wrongness.
Endorsement of utilitarian beliefs, b = -0.153, SE = 0.12, t(161) = -1.276, p = 0.204, η2Part = -0.091, did
not significantly contribute to the model.
4.3 Discussion These results provide additional support for the validity and predictive power of the
Consequentialist scale. In a full factorial model, including deontological and utilitarian beliefs
endorsement of deontological beliefs significantly predicted wrongness ratings. There was no
significant interaction effect between deontological and utilitarian beliefs. Furthermore, deontological
and utilitarian beliefs were not significantly correlated. These findings seem to support the view that
deontological and utilitarian beliefs are independent dimensions.
In this study, the results of Bartels and Pizarro (2011) were replicated; Machiavellianism,
Psychopathy, and Life Meaninglessness were negatively correlated with wrongness ratings. However,
closer inspection of the relationship revealed that only Machiavellianism remained a significant
predictor of wrongness ratings when controlling for Psychopathy and Life Meaninglessness. When
Machiavellianism was added to a model that included deontological and utilitarian beliefs,
endorsement of utilitarian beliefs was no longer a significant predictor of wrongness ratings.
However, endorsement of deontological beliefs remained a significant predictor of wrongness ratings
23
with Machiavellianism included in the model. While the endorsement of utilitarian belief scale did
provide explanatory value over and above Machiavellianism, it remains premature conclude that
Machiavellianism and utilitarianism are synonymous. Although both philosophies believe that the end
justifies the means, for someone who is high in Machiavellianism, the end is self-promotion and
personal gain, while for someone who is high in utilitarianism, the end is the overall welfare of
society. Furthermore, a key tenet of the principled utilitarian position is the realization that personal
suffering should receive no special consideration when making a moral decision. As such, self-
sacrifice is morally obligatory in some situations for the principled utilitarian but would never be
endorsed by a Machiavellian. A fine reading of the utilitarian sub-scale items reveals that none of the
items make this nuanced distinction. This suggests that the endorsement of utilitarian belief sub-scale
should be updated to include items that tap more directly into this prosocial utilitarian ideology.
Furthermore, endorsement of deontological beliefs remained a significant predictor of wrongness
ratings but was not significantly associated with Machiavellianism. This illustrates a complex
relationship between these factors and indicates that perhaps the absence of deontological beliefs is
vital to understanding people’s moral judgments.
24
Chapter 5
5 Politics, Religion and the Consequentialist scale: Rationale
Past research has shown that political beliefs and religiosity play a role in moral judgment.
However, research has failed to outline a clear relationship between these constructs and utilitarian
and deontological thought. For instance conservative political ideology has been associated with
disapproval of gay marriage and abortion (Inbar, Pizarro, & Bloom, 2009), but the rationale for this
association is unclear. It could be due to belief in a utilitarian rationale such that these contentious
social issues threaten the fabric of society and will lead to great misery, or a deontological rationale
that strongly opposes any action that violates the universal conventions of marriage and not killing.
In terms of the influence of religiosity, those high in intrinsic religiosity have been shown to be
more likely to help the homeless (Morgan, Goddard, & Givens, 1997) while at the same time a study
by Vincent, Anker and Feeley (2010) found that 49% of respondents cited religious beliefs as a barrier
to organ donation. Again a conflict arises such that religious beliefs seem to favor a utilitarian
approach in being altruistic, while favoring a more deontological approach when dealing with issues
like organ donation. In short, it is unclear what moral principles individuals are operating with when
making moral judgments even when political orientation and religiosity are known values. Therefore,
this project seeks to demonstrate that the effect of political ideology and religiosity is mediated by
endorsement of deontological and/or utilitarian beliefs.
5.1 Method Participants. A community sample of 157 residents of the United States of America was
collected using Mechanical Turk (operated by Amazon.com). Participants were compensated $0.50
for their participation. The sample consisted of 91 females (57.6%) and 66 males (41.8%). The
average age of participants was 30.65 years (SD = 10.744). Twenty-five percent of the sample was
over the age of 35 years and 50% of the sample was under 29 years of age. Participants reported an
average of 15.47 years of formal education starting from grade one (SD = 3.461). Lastly, the median
25
income range for participants was between $10,000 and $25,000. In terms of income distribution,
54.4% of participants reported earning $25,000 or less in the last calendar year while 19.6% reported
earning $50,000 or more in the last calendar year.
Procedure. Participants logged on to Mechanical Turk and were randomly assigned to one of
three different surveys 1) reactivity, 2) social interaction and 3) higher order beliefs. The data to be
discussed here is from the higher order beliefs survey. Within each survey participants completed a
demographics form and the Consequentialist scale, and read and responded to the same six utilitarian
dilemmas used in chapter 2, section 3.1, and chapter 4. Wrongness ratings across the six dilemmas
were shown to be reliable, α = 0.735, and therefore an aggregate wrongness rating variable was
created for subsequent analysis. As for the Consequentialist scale, both the endorsement of
deontological beliefs sub-scale, α = 0.648, and the endorsement of utilitarian beliefs sub-scale, α =
0.806, were shown to be reliable. In order to test our specific hypothesis participants also completed
the Liberalism scale (Goldberg, 1999) in order to assess political orientation and The Age Universal I-
E Scale (Maltby, 2002) that measures intrinsic religiosity.
Political orientation. Participants’ endorsement of politically conservative ideology was
measured using the Liberalism scale (Goldberg, 1999). This scale contains ten statements (i.e. “I
believe that too much tax money goes to support artists”; “I believe that we coddle criminals too
much”). Participants were asked to indicate endorsement of each statement using a 5-point Likert
scale (where 1 = Strongly Disagree and 5 = Strongly Agree). Higher scores on this scale indicate a
greater propensity for conservative political beliefs. The ten items were shown to be highly reliable, α
= 0.797.
Religiosity scale. One’s orientation towards religion has been conceptualized as being
reducible to two distinct dimensions. Intrinsic religiosity (or intrinsic orientation) is defined as an
individual who views religion as a deeply personal experience. It has been conceptualized as an
individual living their religion and religious beliefs. In contrast, those high in extrinsic religiosity (or
extrinsic orientation) place emphasis on what religion provides for them- for example, protection and
consolation in times of need or a sense of status (Allport & Ross, 1967). Based on the conceptual
differences between intrinsic and extrinsic religiosity it may be hypothesized that those with an
intrinsic religious motivation are more likely to engage with the moral tenants of their faith, and thus
by extension that intrinsic religious motivation will influence their moral beliefs and behavior.
Extending this line of reasoning involved testing whether deontological beliefs predicted moral
judgments over and above intrinsic religiosity. The Age Universal I-E Scale-12 (Maltby, 1999; 2002)
26
was used to assess participants’ religiosity. The scale consists of six statements meant to gauge one’s
intrinsic religiosity (i.e. “It is important to me to spend time in private thought and prayer”).
Participants were asked whether or not the statements described them based on a three-point scale (3 =
yes, 2 = not certain, 1 = no). Endorsement of more items indicated higher intrinsic religiosity. The six
items were shown to be highly reliable, α = 0.873.
5.2 Results
First, I set out to provide additional support for the use of the Consequentialist scale.
Wrongness ratings were regressed on deontological beliefs, utilitarian beliefs, and their interaction
term. The results revealed a significant main effect of deontological beliefs, b = 0.505, SE = 0.126,
t(154) = 3.998, p < 0.0001, η2Part = 0.304, but no significant effect of utilitarian beliefs, b = -0.194, SE
= 0.121, t(154) = -1.607, p = 0.11, η2Part = -0.122, nor a significant interaction, b = -0.159, SE = 0.218,
t(154) = -0.728, p = 0.468, η2Part = -0.055. Endorsement of deontological and utiltiarian beliefs were
not significantly correlated with each other, r(156) = 0.05, p = 0.5.
Subsequent correlational analyses revealed wrongness ratings were significantly correlated
with intrinsic religiosity, r(155) = 0.193, p = 0.015, and endorsement of conservative political values,
r(155) = 0.282, p < 0.0001. Furthermore, endorsement of conservative political beliefs were
significantly correlated with endorsement of deontological beliefs, r(155) = 0.304, p < 0.0001, and
intrinsic religiosity, r(155) = 0.306, p < 0.0001, but not endorsement of utilitarian beliefs, r(155) =
0.054, p = 0.503. Intrinsic religiosity was not significantly correlated with endorsement of utilitarian
beliefs, r(155) = -0.102, p = 0.204.
In order to get a clear picture as to how these variables interact a path analysis was conducted.
The model tested is depicted in Figure 8 (Appendix L). The model appeared to have moderate fit,
χ2(1) = 1.442, p = 0.23, RMSEA = 0.053, 90% CI [0, 0.2], TLI = 0.936.
Endorsement of deontological beliefs predicted wrongness ratings, β = 0.232, Z = 2.928, p =
0.003. Endorsement of utilitarian beliefs had no influence on wrongness ratings, β = -0.117, Z = -
1.573, p = 0.116. Endorsement of conservative political values predicted endorsement of
deontological beliefs, β = 0.211, Z = 2.584, p = 0.01. Endorsement of conservative political values
also retained a significant direct path to wrongness ratings, β = 0.191, Z = 2.312, p = 0.02. This
suggests the influence of the relationship between conservative political values and wrongness
27
judgments is partially mediated by endorsement of deontological beliefs. Conservative ideology had
no significant relationship with endorsement of utilitarian beliefs, β = -0.022, Z = -0.257, p = 0.797.
Intrinsic religiosity predicted endorsement of deontological beliefs, β = 0.219, Z = 2.682, p =
0.007. The direct path between intrinsic religiosity and wrongness ratings was not significant, β =
0.031, Z = 0.372, p = 0.71. This indicates that the influence of intrinsic religiosity on wrongness
ratings was completely mediated by endorsement of deontological beliefs. Intrinsic religiosity had no
relationship with endorsement of utilitarian beliefs, β = -0.092, Z = -1.059, p = 0.29.
Due to the fact that there was no significant relationship between endorsement of utilitarian
beliefs and either endorsement of conservative values or intrinsic religiosity, a second nested model
was tested that constrained the paths between conservative values and endorsement of utilitarian
beliefs, and intrinsic religiosity and utilitarian beliefs, to zero (see Figure 9 – Appendix M). This
model also fit the data well, χ2(3) = 3.135, p = 0.371, RMSEA = 0.017, 90% CI [0, 0.137], TLI = 9.9.
All the paths that were significant in the previous model were also significant in the second model. A
model comparison test revealed that both models fit the data equally well, χ2(2) = 1.693, p = 0.429.
The second model run is more parsimonious and due to the fact both models fit the data equally well
the second model is favoured.
5.3 Discussion
The results above provide additional support for the functionality and validity of the
Consequentialist scale. Deontological beliefs significantly predicted wrongness ratings while
controlling for utilitarian beliefs and any interaction between deontological and utilitarian beliefs.
There was no significant interaction effect between deontological and utilitarian beliefs when
predicting wrongness ratings and deontological beliefs were not significantly correlated with
utilitarian beliefs. These findings seem to support the view that deontological and utilitarian beliefs
are independent dimensions.
Furthermore, a mechanism for how conservative political ideology and intrinsic religiosity is
related to moral judgment was made clear. Both intrinsic religiosity and endorsement of conservative
beliefs were significantly associated with endorsement of deontological beliefs and this in turn
influenced participants’ moral judgments. Interestingly, neither variable was related to endorsement
of utilitarian beliefs. This finding helps to elucidate the judgment pattern of conservatives and
28
religious individuals when it comes to moral matters. This data again supports the view that the
presence of deontological beliefs may be more crucial to understanding moral judgments than the
presence of utilitarian beliefs.
29
Chapter 6
6 Conclusions This thesis first set out to develop a valid and reliable tool to measure individuals’ propensity
towards deontological and utilitarian thought that would not suffer from the limitations of relying
exclusively on moral dilemmas. In chapter one I described and provided evidence for the 10-item
Consequentialist scale. Exploratory factor analysis and two confirmatory factor analyses confirmed a
two-factor structure mapping nicely on the constructs of deontological and utilitarian beliefs. Over the
six studies presented here, the endorsement of deontological beliefs sub-scale was a significant
predictor of wrongness ratings in all six studies (while controlling for endorsement of utilitarian
beliefs and any possible interaction) while the endorsement of utilitarian beliefs sub-scale was a
significant predictor of wrongness ratings in two (chapter 2 – study 1 and chapter 3 – study 2) of the
six studies (while controlling for endorsement of deontological beliefs and any possible interaction).
Furthermore, the Consequentialist scale was shown to be a significant predictor of wrongness ratings
across two different sets of moral dilemmas (chapter 3 study 2). This evidence points to the construct
validity of the Consequentialist scale as it is a potent predictor of moral judgments. One issue that
remains to be clarified is the relationship between deontological and utilitarian beliefs. In chapter one
I found evidence to support the view that deontological beliefs are best thought of as uncorrelated and
interacting while the second CFA that was conducted favored a model in which deontological and
utilitarian beliefs were moderately correlated. In chapters 3 through 5 I found no subsequent evidence
to support the claim that utilitarian and deontological beliefs are correlated. However, I also found no
subsequent evidence to support the idea that these constructs interact. At this point the preponderance
of the evidence seems to suggest that deontological and utilitarian beliefs are best thought of as
independent however further work needs to be done to clarify this relationship.
In chapters 3 through 5 I tested the Consequentialist scale against variables that have been
shown to influence moral judgments. In each of these sections the Consequentialist scale was shown
to provide predictive power over and above the constructs of interest. Furthermore, by examining the
interactions between the sub-scales of the Consequentialist measure a more refined understanding as
to how people make moral judgments has emerged. For instance in chapter three, the relationship
30
between emotional reactivity and wrongness ratings could not be explained by appealing to increased
deontological or utilitarian beliefs. Also, the effect of cognitive reflexivity on wrongness ratings was
explained by a decrease in deontological beliefs. This is counter to previous theory which has
associated emotional reactivity with increase rule-based (deontological thought) and increased
cognitive reflexivity with a higher degree of utilitarian thinking. Chapter four demonstrated that the
influence of utilitarian beliefs on wrongness ratings was explained by the presence of
Machiavellianism. However, endorsement of deontological beliefs continued to be a significant
predictor of wrongness ratings when Machiavellianism was included in the model. Lastly, the
influence of intrinsic religiosity and political conservatism on wrongness ratings was shown to be due
to increased endorsement of deontological beliefs and no relationship with utilitarian beliefs.
In summary, the Consequentialist scale provides short and reliable measure of deontological
and utilitarian beliefs that eliminates the possible confounding factors associated with the use of moral
dilemmas. The Consequentialist scale was shown to have a valid factor structure and internal
reliability while also being predictive of moral judgments. Lastly, the Consequentialist scale was
shown to add predictive and explanatory power over and above constructs previously associated with
moral judgments helping to give us a more refined picture as to the process individuals engage in
when making moral judgments.
31
References
Allport, G.W., & Ross, J.M. (1967). Personal religious orientation and prejudice. Journal of Personality and Social Psychology, 5, 432-433.
Baron, J., & Ritov, I. (2009). Protected values and omission bias as deontological judgments. In D.M. Bartels, C.W. Bauman, L.J. Skitka, & D.L. Medin (Eds.). Moral judgment and decision- making: The psychology of learning and motivation. (Vol. 50, pp.133-167). San Diego: Elsevier.
Bartels, D. M., & Pizarro, D. A. (2011). The mismeasure of morals: antisocial personality traits predict utilitarian responses to moral dilemmas. Cognition, 121(1), 154-61.
Bartels, D.M. (2008). Principled moral sentiment and the flexibility of moral judgment and decision-making. Cognition, 108, 381-417. Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108, 624 – 652.
Cannon, P. R., Schnall, S., & White, M. (2010). Transgressions and Expressions: Affective Facial Muscle Activity Predicts Moral Judgments. Social Psychological and Personality Science, 2(3), 325-331.
Casali, G. L. (2011). Developing a Multidimensional Scale for Ethical Decision Making. Journal of Business Ethics, 104(4), 485-497.
Chapman, H. A. Kim, D. A, Susskind, J. M., & Anderson, A K. (2009). In bad taste: evidence for the oral origins of moral disgust. Science, 323, 1222-6.
Foot, P. (1967). The Problem of Abortion and the Doctrine of Double Effect. Oxford Review, 5, 5–15
Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. In I. Mervielde, I. J. Deary, F. De Fruyt, & F. Ostendorf (Eds.), Personality psychology in Europe (Vol. 7, pp. 7-28). Tilburg, Netherlands: Tilburg University Press.
Greene, J.D., Cushman, F.A., Stewart, L.E., Lowenberg, K., Nystrom, L.E., & Cohen, J.D. (2009) Pushing moral buttons: The interaction between personal force and intention in moral judgment. Cognition, 111, 364-371 Greene, J. D., Morelli, S. A., Lowenberg, K., Nystrom, L. E., & Cohen, J. D. (2008a). Cognitive load selectively interferes with utilitarian moral judgment. Cognition, 107, 1144 –1154.
32
Greene, J. D. (2008b). The secret joke of Kant’s soul. In W. Sinnott -Armstrong (Ed.), Moral Psychology: Vol. 2. The cognitive science of morality (pp. 35 - 79). Cambridge, MA: MIT Press. Greene, J. D., Nystrom, L. E., Engell, A. D., Darley, J. M., & Cohen, J. D. (2004). The neural bases of cognitive conflict and control in moral judgment. Neuron, 44, 389 – 400. Greene, J. D., Sommerville, R. B., Nystrom, L. E., Darley, J. M., & Cohen, J. D. (2001). An fMRI study of emotional engagement in moral judgment. Science, 293, 2105 – 2108.
Haidt, J. (2001). The Emotional Dog and Its Rational Tail : A Social Intuitionist Approach to Moral Judgment, Psychological Review, 108(4), 814-834.
Helzer, E. G., & Pizarro, D. a. (2011). Dirty liberals! Reminders of physical cleanliness influence moral and political attitudes. Psychological science, 22(4), 517-22.
Inbar, Y., Pizarro, D. A, Knobe, J., & Bloom, P. (2009). Disgust sensitivity predicts intuitive disapproval of gays. Emotion, 9, 435-439.
Kant, I. (1959). Foundation of the metaphysics of morals (L. W. Beck, Trans.). Indianapolis: Bobbs - Merrill. (Original work published in 1785.) Koenigs, M., Young, L., Adolphs, R., Tranel, D., Cushman, F., Hauser, M., Damasio, A. (2007). Damage to the prefrontal cortex increases utilitarian moral judgments. Nature, 446, 908 – 911. Kruglanski, A.W., Webster, D.M., & Klem, A. (1993). Motivated resistance and openness to persuasion in the presence or absence of prior information. Journal of Personality and Social Psychology, 65(5), 861-876. Kunzendorf, R.G., Moran, C., & Gray, R. (1996). Personality traits and reality-testing abilities, controlling for vividness of imagery, Imagination, Cognition and Personality, 15, 113-131. Lind, G. (1978). How does one measure moral judgment? Problems and alternative ways of measuring a complex construct. In G. Portele (Ed.), Socialization and moral (pp. 171–201). Basel: Beltz.
Maltby, J. (2002). The Age Universal I-E Scale-12 and Orientation Toward Religion : Confirmatory Factor Analysis. The Journal of Psychology, 136(5), 555-560.
Maltby, J. (1999). The internal structure of a derived, revised, and amended measure of the religious orientation scale: The “age-universal” I-E Scale - 12. Social Behavior and Personality, 27(4), 407-412.
Mill, J.S. (1861/1998). Utilitarianism. In R. Crisp (Ed.), New York: Oxford University Press.
33
Moore, A. B., Clark, B. A., & Kane, M. J. (2008). Who Shalt Not Kill ? Individual Differences in Working Memory Capacity, Executive Control, and Moral Judgment. Psychological Science, 19(6), 549-557.
Morgan, M. M., Goddard, H. W., & Givens, S. N. (1997). Factors That Influence Willingness to Help the Homeless. Journal of Social Distress and the Homeless, 6(1), 45-56.
Paulhus, D. L., & Williams, K. M. (2002). The Dark Triad of personality: Narcissism, Machiavellianism, and psychopathy. Journal of Research in Personality, 36(6), 556-563.
Paxton, J. M., Ungar, L., & Greene, J. D. (2011). Reflection and reasoning in moral judgment. Cognitive science, 36(1), 163-77.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879-891.
Reidenbach, R. E., & Robin, D. P. (1988). Some initial steps toward improving the measurement of ethical evaluations of marketing activities. Journal of Business Ethics, 7(11), 871–879.
Rest, J. R., Narvaez, D., Thoma, S. J., & Bebeau, M. J. (1999). DIT2: Devising and testing a revised instrument of moral judgment. Journal of Educational Psychology, 91(4), 644-659.
Rudd, A., Mullane, S., & Stoll, S. (2010). Development of an Instrument to Measure the Moral Judgments of Sport Managers. Recreation, 59-82.
Schnall, S., Haidt, J., Clore, G. L., & Jordan, A. H. (2008). Disgust as embodied moral judgment. Personality & social psychology bulletin, 34(8), 1096-109.
Stevens, J. (1996). Applied multivariate statistics for the social sciences (3rd: Edition). Mahwha, New Jersey: Erlbaum.
Sunstein, C.R. (2005). Moral heuristics. Behavioral and Brain Sciences, 28, 531-542
Suter, R. S., & Hertwig, R. (2011). Time and moral judgment. Cognition, 119(3), 454-8.
Terrizzi, J. a., Shook, N. J., & Ventis, W. L. (2010). Disgust: A predictor of social conservatism and prejudicial attitudes toward homosexuals. Personality and Individual Differences, 49(6), 587- 592.
Thomson, J. (1985). The trolley problem. Yale Law Journal, 94(6), 1395–1415.
Tybur, J. M., Lieberman, D., & Griskevicius, V. (2009). Microbes, mating, and morality: individual differences in three functional domains of disgust. Journal of personality and social psychology, 97(1), 103-22.
34
Vincent, D. E., Anker, A. E., & Feeley, T. H. (2011). Religion and the Decision to Donate Organs : Exploring the Potential Role of Religious Leaders, Journal of Community & Applied Social Psychology, 21, 312-328.
Wheatley, T., & Haidt, J. (2005). Hypnotic Disgust Makes Moral Judgments More Severe. Psychological science, 16(10), 780-784.
Zgheib, P. W. (2005). Managerial Ethics: An Empirical Study of Business Students in the American University of Beirut. Journal of Business Ethics, 61(1), 69-78.
35
Appendices
Appendix A.
Table 1. Dilemma Summary
Original Name Dilemmas as they appeared to participants. % Utilitarian Judgment
Vitamin Deficiency
You are the leader of a mountaineering expedition that is stranded in the wilderness. Your expedition includes a family of six that has a genetically caused vitamin deficiency. A few people’s kidneys contain large amounts of this vitamin. There is one such person in your party. The only way to save the lives of the six members of this family is to remove one of this man’s kidneys so that the necessary vitamins may be extracted from it. The man will not die if you do this, but his health will be compromised. The man is opposed to this plan, but you have the power to do as you see fit. How wrong would it be for you to forcibly remove the man's kidney in order to save the lives of the six vitamin-deficient people?
35%
Modified Life Boat
You are on a cruise ship when there is a fire on board, and the ship has to be abandoned. The lifeboats are carrying many more people than they were designed to carry. The lifeboat you’re in is sitting dangerously low in the water—a few inches lower and it will sink. The seas start to get rough, and the boat begins to fill with water. If nothing is done it will sink before the rescue boats arrive and everyone on board will die. However, there is an injured person who will not survive in any case. If you throw that person overboard the boat will stay afloat and the remaining passengers will be saved. How wrong would it be for you to throw the injured person overboard in order to save the lives of the remaining passengers on the lifeboat?
71%
Sacrifice
You, your spouse, and your four children are crossing a mountain range on your return journey to your homeland. You have inadvertently set up camp on a local clan’s sacred burial ground. The leader of the clan says that according to the local laws, you and your family must be put to death. However, he will let yourself, your spouse, and your three other children live if you yourself will kill your oldest child. How wrong would it be for you to kill your oldest child in order to save your spouse and your other three children?
51%
36
Sophie's Choice
It is wartime and you and your two children, ages eight and five, are living in a territory that has been occupied by the enemy. At the enemy’s headquarters is a doctor who performs painful experiments on humans that inevitably lead to death. He intends to perform experiments on one of your children, but he will allow you to choose which of your children will be experimented upon. You have twenty-four hours to bring one of your children to his laboratory. If you refuse to bring one of your children to his laboratory he will find them both and experiment on both of them. How wrong would it be for you to bring one of your children to the laboratory in order to avoid having both of them die?
62%
Euthanasia
You are the leader of a small group of soldiers. You are on your way back from a completed mission deep in enemy territory when one of your men has stepped in a trap that has been set by the enemy and is badly injured. The trap is connected to a radio device that by now has alerted the enemy to your presence. They will soon be on their way. If the enemy finds your injured man they will torture him and kill him. He begs you not to leave him behind, but if you try to take him with you your entire group will be captured. The only way to prevent this injured soldier from being tortured is to shoot him yourself. How wrong would it be for you to shoot this soldier in order to prevent him from being tortured by the enemy?
63%
Crying Baby
Enemy soldiers have taken over your village. They have orders to kill all remaining civilians. You and some of your townspeople have sought refuge in the cellar of a large house. Outside you hear the voices of soldiers who have come to search the house for valuables. Your baby begins to cry loudly. You cover his mouth to block the sound. If you remove your hand from his mouth his crying will summon the attention of the soldiers who will kill you, your child, and the others hiding out in the cellar. To save yourself and the others you must smother your child to death. How wrong would it be for you to smother your child in order to save yourself and the other townspeople?
60%
37
Appendix B.
Table 2. Pattern Matrix - Exploratory Factor Analysis
Item Code Item Factor 1 Factor2
Deont_1 Some rules should never be broken. 0.039 0.502
Deont_2 It is never morally justified to cause someone harm. 0.145 0.642
Deont_3 The consequences of an action are not the only factor in deciding if an action is moral; even if the consequences of the action are positive. -0.297 0.109
Deont_4 If an action is a violation of societies most basic rules it should not be committed; even if it will result in a large amount of good. 0.117 0.559
Deont_5 Some aspects of humanity are sacred and should never be violated no matter the possible gain. -0.092 0.534
Deont_6 It is important that one abide by the rules of society. -0.092 0.314
Deont_7 Killing is never justified. 0.174 0.561
Deont_8 One should always tell the truth. 0.004 0.389
Deont_9 It is important to only perform actions that are honorable. 0.028 0.331
Deont_10 Certain rights and freedoms are inalienable and should never be violated. -0.1 0.244
Deont_11 The results of an action are not enough to justify that action. -0.16 0.305
Deont_12 Happiness is not the only thing we should strive for; we should seek knowledge and try to live a productive life even if that means not always being happy. -0.323 -0.024
Deont_13 It is morally unacceptable to consider only the good that might come from an action when deciding how to behave. -0.127 0.018
Deont_14 You should never treat people in a manner where they are means to a greater end. -0.123 0.291
Deont_15 Sometimes the morally correct thing to do causes more pain then happiness. -0.226 0.034
Deont_16 When deciding if an action is moral or immoral people have to take several dimensions into account and weigh these dimensions out carefully. -0.168 -0.103
Deont_17 It is not morally acceptable to correct one violation of the moral code with another violation of the moral code. -0.115 0.285
Deont_18 Some rules and laws are universal and are binding no matter the circumstances you find yourself in. 0.043 0.495
Deont_19 Violating rules and laws that are universal is always wrong. 0.164 0.55
Deont_20 If you could make the world a better place by violating a universal rule or law the correct thing to do would be to do nothing. 0.125 0.371
Utili_1 Causing someone harm is the morally correct thing to do if it leads to some greater good. 0.338 -0.369
Utili_2 Moral actions are the ones that create the most good; even if the act itself causes harm or pain initially. 0.236 -0.041
Utili_3 The needs of the many always outweigh the needs of the few or the individual. 0.331 -0.106
Utili_4 When deciding what action to take the only relevant factor to consider is the outcome of the action. 0.523 0.012
Utili_5 If an action leads to a greater good then you should do it. 0.28 -0.145
Utili_6 Rules and laws are irrelevant; whether an action produces happiness is all that matters when deciding how to act. 0.621 -0.057
Utili_7 It is acceptable to kill as long as that death leads to some greater good. 0.307 -0.407
Utili_8 Lying is wrong unless it leads to generating a greater amount of happiness. 0.325 -0.148
Utili_9 Rules and laws should only be followed when they maximize happiness. 0.68 -0.033
Utili_10 If rules and laws do not maximize happiness for people they should be ignored. 0.631 -0.032
38
Utili_11 No rule or law is sacred; they can be broken in order to serve some greater good. 0.133 -0.413
Utili_12 Whether one has maximized happiness is the only standard by which one should measure the success of one's life. 0.578 0.238
Utili_13 Saving many lives by killing one person is morally acceptable. 0.25 -0.397
Utili_14 Allowing people to experience pain and suffering in the name of maintaining a set of principles is morally unacceptable. 0.123 0.052
Utili_15
Absolute moral codes are an illusion: when deciding how to behave one must only consider whether their actions will have positive or negative consequences.
0.334 -0.213
Utili_16 Every rule or law has a circumstance where it is morally acceptable to act contrary to it. 0.245 -0.171
Utili_17 If performing an action is going to cause you pain but create larger amount of good then you should do it. 0.141 -0.162
Utili_18 The only moral principle that needs to be followed is that one must maximize happiness. 0.711 0.141
Utili_19 People that fail to maximize happiness are doing something morally wrong. 0.595 0.133
Utili_20 The result of one's actions is all that matters when deciding if they have acted morally. 0.531 0.055
**Bolded items were those retained for subsequent analysis.
39
Appendix C.
Figure 1. Scree Plot – Exploratory factor analysis
!
40
Appendix D.
Table 3. Pattern Matrix - Refined Exploratory Factor Analysis
Item Code Items Factor 1 Factor 2
Deont_1 Some rules should never be broken. -0.033 0.582
Deont_2 It is never morally justified to cause someone harm. 0.147 0.549
Deont_4 If an action is a violation of societies most basic rules it should not be committed; even if it will result in a large amount of good. 0.073 0.553
Deont_5 Some aspects of humanity are sacred and should never be violated no matter the possible gain. -0.121 0.592
Deont_7 Killing is never justified. 0.181 0.452
Deont_18 Some rules and laws are universal and are binding no matter the circumstances you find yourself in. -0.014 0.577
Utili_4 When deciding what action to take the only relevant factor to consider is the outcome of the action. 0.47 0.028
Utili_6 Rules and laws are irrelevant; whether an action produces happiness is all that matters when deciding how to act. 0.659 -0.152
Utili_9 Rules and laws should only be followed when they maximize happiness. 0.712 -0.114
Utili_10 If rules and laws do not maximize happiness for people they should be ignored. 0.676 -0.111
Utili_11 No rule or law is sacred; they can be broken in order to serve some greater good. 0.159 -0.4
Utili_12 Whether one has maximized happiness is the only standard by which one should measure the success of one's life. 0.563 0.178
Utili_18 The only moral principle that needs to be followed is that one must maximize happiness. 0.71 0.063
Utili_19 People that fail to maximize happiness are doing something morally wrong. 0.582 0.106
Utili_20 The result of one's actions is all that matters when deciding if they have acted morally. 0.484 0.057
**Items in bold are those retained for confirmatory factor analysis
41
Appendix E.
Figure 2. Preliminary confirmatory factor analysis 1
!
42
Appendix F
Figure 3. Preliminary confirmatory factor analysis model 2
!
43
Appendix G.
Figure 4a. Consequentialist Belief Interaction CFA 1
Figure 4b. Consequentialist Belief Interaction CFA 2
3
3.5
4
4.5
5
5.5
6
6.5
7
Low Deontology High Deontology
Wrongness Ratings
Deontological Beliefs
Consequentialist Belief Interaction CFA 1
Low Utilitarian
High Utilitarian
3
3.5
4
4.5
5
5.5
6
6.5
7
Low Deontology High Deontology
Wrongness Ratings
Deontological Beliefs
Consequentialist Belief Interaction CFA 2
Low Utilitarian
High Utilitarian
44
Appendix H.
Figure 5. Second confirmatory factor analysis model 1
!
45
Appendix I.
Figure 6. Second confirmatory factor analysis model 2
!
46
Appendix J.
Table 4. Moore and Colleagues Dilemma Summary
Original Name Dilemmas as they appeared to participants. % Utilitarian Judgment
Mine Shaft
You are a worker in a mine. The only way to exit the mine is to ride up in rock buckets that can hold up to three people at a time. It is the end of the workday and miners from lower levels are riding up past you. As you are waiting for a bucket to ride in, you notice that the cable supporting the rock buckets is about to snap. If the cable snaps, all of the miners in the buckets will fall to their deaths. The only way to prevent this is to use your axe to hit the last bucket causing it to flip over and dump its contents, lightening the load enough to save the miners above. There is one miner in this bucket who will be killed as a result.
44%
Modified Vaccine
A viral epidemic has spread across the globe killing thousands of people. You are a medical researcher and have developed two substances in your laboratory. You know that one of them is a vaccine, but you don’t know which because both of the vials have been mislabeled as vaccine. You also know that the other substance is deadly. Once you figure out which substance is the vaccine you can create more to save thousands of lives. You have two lab assistants who work with you, and the only way to identify the vaccine with certainty is to inject the two substances into these people against their wishes. One person will live, the other will die, and you will be able to start saving lives with your vaccine.
56%
Waterfront
You are part of a shipyard dock team that attaches crane cables to huge transport containers that are to be unloaded from ships onto the shore. You and the others attach these cables and then ride on top of the containers, wearing safety harnesses, to make sure that the containers are unloaded properly. While you are riding on top of one container that is just being lifted out of the cargo bay, you see the red warning light that indicates that the crane cable is about to fail. You realize that if the cable fails and the container falls onto the deck of the ship, many of the crewmembers below will be crushed to death. You see that the cable is in danger because two other crewmembers are fighting on top of the container, causing it to sway dangerously. You could run over to the fight and push them apart, but one of the crewmembers has taken off his safety harness, and will certainly fall to his death. However, if you do not do this, the continued swaying will cause the cable to fail and the container to fall onto the deck, killing several people below.
56%
47
Bus Plunge
You are the bus driver for a kindergarten field trip to the zoo. On the way, faulty tires cause the bus to overturn and plunge off of a bridge and intoa roaring river. You and three of the children are still in the sinking bus, but the rest of the passengers have been swept away down the river to theirdeaths. You grab the two children nearest to you and begin to swim toward the exit door. The third remaining child grabs onto your leg. You realize that you are not strong enough to fight the current and swim with all three children holding on to you. The only way to reach thesurface before you and the children drown is to shake the third child off of your leg. This will allow you to bring the two children to the surface with you, but the third child will drown.
49%
Cinderblock
You are the explosives expert for a company that has been hired to demolish a skyscraper. You are examining the last of the explosive charges when you notice a teenager below who is about to accidentally detonate one of the charges out of sequence. This explosion will result in the building’s uncontrolled collapse onto you, the teenager, and the crowd of spectators. The teenager is several floors below you and cannot hear you because of the loud demolition noise. You realize that the only way to stop the teenager from detonating the charge is to drop a heavy cinderblock on his head. This will crush his skull and kill him almost instantly but will prevent the out of sequence explosion.
51%
Nuclear Reactor
You are the inspector of a nuclear power plant that you suspect has not met its safety requirements. The plant foreman and you are touring the facility when one of the nuclear fuel rods overheats. The emergency coolant system fails to activate, and a chain reaction is about to begin which will result in a nuclear meltdown. This will release lethal radiation into the nearby town, killing thousands of people. You realize that the only way to stop the meltdown is to push the foreman into the fuel rod assembly. This will remove just enough heat energy from the rod assembly to prevent the nuclear chain reaction. However, it will also incinerate the foreman instantly.
50%
Rescue 911
You are the sole paramedic riding on a rescue helicopter, responding to a devastating flood. You have rescued several critically injured people and you have been treating them to keep them alive. Suddenly, the pilot notifies you that there has been an engine malfunction and the helicopter can no longer support the weight of you, the pilot, and all the injured people on board. If the helicopter crashes, it will kill everyone on board including you, the pilot, and the injured people that you have rescued. In order to avoid a crash that will kill everyone on the helicopter, you realize that you must lighten the load enough to keep the helicopter aloft. Because all of the equipment is bolted down, and you are needed by the injured people, the only way to do this is to throw one of the injured people off the helicopter. This will lighten the load enough to keep the helicopter in the air but will cause the death of the person thrown off.
45.70%
48
Appendix K.
Figure 7. Cognitive reflection, Deontological Beliefs and Wrongness Ratings: Mediation Analysis
!"#$%&'()*(+(,&"$)
-("$."/"#%,0/)1(/%(23)
45"$#$(33)*0&$#3)67879:);678<=><?@)
?A)B787>:)??CB787D:)???AB7877D)
678D9EF???) 789DG<???)
H%#I5()G)
49
Appendix L.
Figure 8. Affect of Religion, Politics, and Consequentialist Beliefs: Model 1
!"#$%"&%'()*+%,%-&%#.(
/-"&*$0120*(3-+%2'1+(41+5*&(
6$-","*&&()12",&(
78(9:;:<=(7739:;:>=(77789:;::>(
?:;>@A(
B%,5$*(@(
C*-"#-+-,%'1+(D*+%*E&(
F2+%#1$%1"(D*+%*E&(
:;G@A77(
:;AHI77(
?:;:AH(
:;AJ>77(
:;>JH777(
?:;>AJ(:;:K<(
:;GK7(
50
Appendix M.
Figure 9. Affect of Religion, Politics, and Consequentialist Beliefs: Model 2
!"#$%"&%'()*+%,%-&%#.(
/-"&*$0120*(3-+%2'1+(41+5*&(
6$-","*&&()12",&(
78(9:;:<=(7739:;:>=(77789:;::>(
?:;>@A(
B%,5$*(C((
D*-"#-+-,%'1+(E*+%*F&(
G2+%#1$%1"(E*+%*F&(
:;H@A77(
:;AIC77(
:;AJ>77(
:;>JI777(
:;:K<(
:;HK7(
51
Appendix N: Table 5. Model Summary Section 3.4
Unstandardized Coefficients
Model b Std. Error t Sig
Constant 4.247 0.111 38.407 <0.0001
Cognitive Reflect Test (CRT) -‐0.109 0.096 -‐1.129 0.26
Deontological Beliefs 0.786 0.154 5.1 <0.0001
Utilitarian Beliefs -‐0.328 0.146 -‐2.247 0.026
CRT*Deontology -‐0.145 0.13 -‐1.115 0.266
CRT*Utility -‐0.169 0.125 -‐1.354 0.177
Deontology*Utility -‐0.075 0.205 -‐0.367 0.714
Deontology*Utility*CRT -‐0.093 0.157 -‐0.591 0.555
52
Appendix O: Table 6. Correlation Summary Section 3.4
1 2 3 4
1. Wrongness Ratings 1
2. Deontological Beliefs 0.381*** 1
3. Utilitarian Beliefs -‐0.165* -‐0.101 1
4. Cognitive Reflection
Task (CRT)
-‐0.166* -‐0.3111** 0.017 1
p < 0.05*, p <0.01**, p < 0.001***