1. Descriptive Statistics: Produce a descriptive statistics table for all numerical variables, includingthe mean, standard deviation, max value, and min value.
2. Correlation table: In Jamovi or your preferred analysis software, reproduce the correlation table
for Scenario 2 (page 5, labels do not need to be an exact match).
3. T-tests: Use Independent Sample T-tests to replicate the Scenario 2 results for the five
dependent variables by omission vs commission conditions. Include effect size, confidence
intervals, and descriptive plots.
4. ANOVA: The ANOVA test of the interaction effect of omission and no-harm outcome on
immorality in Scenario 2. Include a plot generated by the Estimated Marginal Means to show the
confidence intervals.
Deliverables:
1. .omv file: When you’ve run your analyses and you see the data replicated, save your file in
Jamovi as an .omv file and submit that file to this assignment here. If you use another analysis
software, submit your script here (e.g., .r, .rmd, .spv) as well as a Word document with the
output tables copy/pasted in. Your tables don’t need to look exactly like the ones published, but
your numbers must match the results reported in the paper.
1. Descriptive Statistics: Produce a descriptive statistics table for all numerical variables, including
the mean, standard deviation, max value, and min value.
2. Correlation table: In Jamovi or your preferred analysis software, reproduce the correlation table
for Scenario 2 (page 5, labels do not need to be an exact match).
3. T-tests: Use Independent Sample T-tests to replicate the Scenario 2 results for the five
dependent variables by omission vs commission conditions. Include effect size, confidence
intervals, and descriptive plots.
4. ANOVA: The ANOVA test of the interaction effect of omission and no-harm outcome on
immorality in Scenario 2. Include a plot generated by the Estimated Marginal Means to show the
confidence intervals.
Deliverables:
1. .omv file: When you’ve run your analyses and you see the data replicated, save your file in
Jamovi as an .omv file and submit that file to this assignment here. If you use another analysis
software, submit your script here (e.g., .r, .rmd, .spv) as well as a Word document with the
output tables copy/pasted in. Your tables don’t need to look exactly like the ones published, but
your numbers must match the results reported in the paper.
Journal of Experimental Social Psychology 89 (2020) 103977
Contents lists available at ScienceDirect
Journal of Experimental Social Psychology
journal homepage: www.elsevier.com/locate/jesp
Case Report
Action-inaction asymmetries in moral scenarios: Replication of the omission
bias examining morality and blame with extensions linking to causality,
intent, and regret☆
T
John Jamisona,1, Tijen Yayb,1, Gilad Feldmanb,c, ,1
⁎
a
Department of Management, Hong Kong University of Science and Technology, Hong Kong Special Administrative Region
Department of Work and Social Psychology, Maastricht University, Maastricht 6200MD, the Netherlands
c
Department of Psychology, University of Hong Kong, Hong Kong Special Administrative Region
b
ARTICLE INFO
ABSTRACT
Keywords:
Omission bias
Omission
Commission
Action
Inaction
Morality
Blame
Attributions
Omission bias is the preference for harm caused through omissions over harm caused through commissions. In a preregistered experiment (N = 313), we successfully replicated an experiment from Spranca, Minsk, and Baron (1991),
considered a classic demonstration of the omission bias, examining generalizability to a between-subject design with
extensions examining causality, intent, and regret. Participants in the harm through commission condition(s) rated
harm as more immoral and attributed higher responsibility compared to participants in the harm through omission
condition (d = 0.45 to 0.47 and d = 0.40 to 0.53). An omission-commission asymmetry was also found for perceptions
of causality and intent, in that commissions were attributed stronger action-outcome links and higher intentionality
(d = 0.21 to 0.58). The effect for regret was opposite from the classic findings on the action-effect, with higher regret
for inaction over action (d = −0.26 to −0.19). Overall, higher perceived causality and intent were associated with
higher attributed immorality and responsibility, and with lower perceived regret. All materials are available on:
https://osf.io/9gsqe/
1. Introduction
Omission bias is a well-researched phenomenon of the preference for
harm caused through omission over harm caused through commission, even
when the outcome is the same (Baron and Hershey, 1988; Baron & Ritov,
1994; Connolly & Reb, 2003; Kordes-de Vaal, 1996). People facing moral
dilemmas between taking action or doing nothing when outcomes are unknown and/or uncertain typically choose not to act, even when both decisions are perceived as likely to lead to similar negative outcomes. The present investigation aims to closely replicate a classic finding of the omission
bias employing a procedural adjustment and extensions to the original design to address open questions and key developments in the literature.
1.1. Omission bias
Early research into omission bias shed light on the bias of parents
towards inaction in vaccination decisions, despite the likelihood of such
decisions resulting in negative outcomes (Ritov & Baron, 1990, 1992).
Data from the United States in the early 1990s showed puzzling lower
than expected vaccination rates. Ritov and Baron argued that this could
be explained by an omission bias, such that parents perceived a decision
for risking death from the flu by not vaccinating their child (omission)
as being less immoral and incurring less responsibility than risking
death from side effects of vaccinating their child (commission) (Ritov &
Baron, 1990). Their findings showed that despite both decisions resulting in the same or even greater chances or degree of a harmful
outcome, there was a clear preference for decisions of omission over
commission. Asch et al. (1994) extended these findings to show that
these lab experiments generalized to real vaccination decisions, with
later studies finding support for an omission bias in many other moral
and non-moral domains (Cushman & Young, 2011; Meszaros et al.,
1996).
This paper has been recommended for acceptance by Robbie Sutton.
Corresponding author at: Department of Psychology, University of Hong Kong, Hong Kong Special Administrative Region.
E-mail addresses: jmjamison@ust.hk (J. Jamison), t.yay@student.maastrichtuniversity.nl (T. Yay), gfeldman@hku.hk,
gilad.feldman@maastrichtuniversity.nl (G. Feldman).
1
Contributed equally, joint first author.
☆
⁎
https://doi.org/10.1016/j.jesp.2020.103977
Received 8 July 2018; Received in revised form 5 March 2020; Accepted 8 March 2020
Available online 07 May 2020
0022-1031/ © 2020 Elsevier Inc. All rights reserved.
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
Despite accumulating evidence in support of the omission bias,
there were scholars, such as Connolly and Reb (2003), who challenged
earlier findings and questioned the generalizability of the bias from the
context of vaccination decisions, making the case that the evidence for a
general omission bias is weak and inconclusive. The criticisms of a
generalizable omission bias have tended to focus on two issues. First,
regarding the scale of the moral decision being examined, it has been
argued that the vaccination decision is an overly complex social-moral
dilemma that is not a suitable test for the omission bias (Reb &
Connolly, 2010). This includes criticisms that decisions of life and death
may be tied to unflexing beliefs where alternatives involving some risk
of death or serious side effects like those involved with vaccines might
not be considered. Second, there have also been concerns regarding
measurement and study design. The first of these was that the differentiation between omission and commission strongly depends on the
response scales used in vaccination studies that were not themselves
designed with this kind of rigorous academic analysis in mind (e.g. yes/
no questionnaires in Asch et al., 1994). The next measurement and
design concern has been the use of a numerical risk-balancing procedure in which participants have to make tradeoffs of probabilities to
vaccinate or not (Hershey, Asch, Thumasathit, Meszaros, & Waters,
1994; Meszaros et al., 1996; Petrinovitch & O’Neill, 1996). These were
seen as problematic in that it may be more straightforward and accurate
to ask people to make a simple choice between ‘vaccinate’ and ‘not
vaccinate’ rather than to address complicated computational evaluations with varying probabilities (Connolly & Reb, 2003; Reb &
Connolly, 2010). Finally, some have postulated that the within-subject
studies do not replicate well when conducted in a between–subject
design (e.g. Connolly & Reb, 2003; N’gbala & Branscombe, 1997). Together, these issues, the morally complex context of vaccination, and
the measurement and design of early studies, have raised the question
of the generalizability of the omission bias from the particular case of
vaccinations to a broader human condition.
bias. Firstly, it was not clear whether the results would generalize to a
between-subjects design in which participants would judge each case
based on its own merits rather than in comparison with other cases.
Between-subjects designs also help address concerns of subjects’
awareness of the manipulation and adjustments of their answers by
making comparisons across conditions. Baron and Ritov (2004) acknowledge that these contrasts, if real, would be interesting to explore,
though they speculate that the omission bias is not likely to disappear
when subjects are aware of it. The authors insisted in follow-up studies
that this within-subject design is important for holding intentionality
constant (e.g. Baron & Ritov, 2009; Royzman & Baron, 2002), though
the variation of intentionality is perhaps one interesting boundary
condition on the omission bias effect (Kordes-de Vaal, 1996).
Additionally, recent large-scale mass collaborations have shown
alarmingly high rates of failure to replicate classic findings in psychology (e.g. Klein et al., 2018) leading to a “credibility revolution” and
increasing calls for more replication work (Zwaan, Etz, Lucas, &
Donnellan, 2018). These serve as an opportunity to revisit classic
findings on the omission bias, to further extend these findings to allow a
better understanding of the phenomenon, and to address questions
raised regarding the existence and the strength of the effect.
In the present investigation, we aimed to test the generalizability
and extension of the omission bias by closely replicating and extending
the 1991 study. We sought first to replicate the study with a procedural
adjustment to a between-subjects design to test its generalizability, and
then to extend the study with additional outcomes of participant perceptions of causality, intentionality, and regret. For the purposes of
replication, we focused on their first experiment which consisted of two
scenarios and served as the baseline for the other follow-up experiments. We aimed to replicate the original study as closely as possible,
using their original scenarios and outcomes as source material. As far as
we know, there have been no previous attempts to conduct a direct
replication of this experiment.
We aimed for a close replication according to the replication taxonomy proposed by LeBel, McCarthy, Earp, Elson, and Vanpaemel
(2018). We made adjustments to the original within-person design in
which subjects read and evaluated all conditions of the same scenario,
both commission(s) and omission, and we instead presented participants with either commission or omission. This was meant to address
one of the main criticism of the omission bias findings in the ongoing
debate in the literature was that the effect relies on within-subject or
single-choice designs. Therefore, we chose a well-known demonstration
of the omission-bias using a within-subject design and attempted a close
direct replication using the same stimuli with a between-subject design.
We also made minor adjustments to validate participants’ understanding of action and inaction in the scenarios. We added comprehension checks to ensure participants read and understood the scenarios in the way that was intended, thereby also disrupting automated
or random responding and forcing participants to pay attention to key
factors in the scenarios, and addressing any possible concerns that
participants may have not understood the distinction between commission and omission in the scenarios.
1.2. Chosen omission bias experiment
Spranca, Minsk, and Baron (1991) provided one of the first experimental accounts of a broad omission bias, with five experiments
demonstrating the effect and exploring possible explanations. Their
study has become a classic in the omission bias literature with 805 citations according to Google Scholar at the time of writing, and 766 of
these citations including references to action and inaction.2 Their
study’s experiments first sought to address the moral complexity criticism by presenting participants with simple and straightforward everyday life examples of the bias, avoiding the political and cultural
complexities of the vaccination dilemma scenarios that are at the center
of the omission bias debate. Their experiments also sought to address
the criticism of complex measurement method by removing the complexities of statistical and probabilistic evaluations in preference for
more familiar ratings of morality on a scale of 0 to −100 and of responsibility as a reward to the victim. These steps made a strong case
that the omission bias should be observed in less dire moral contexts
and with more straightforward measurement techniques. This study
along with others (e.g., Baron & Ritov, 2004, 2009) made steps to address challenges to the generalizability of the omission bias, yet the
debate continues (e.g., Connolly & Reb, 2012a, 2012b; Willemsen &
Reuter, 2016).
1.4. Extensions: causality, intentionality, and regret
In addition to the purpose of replicating the 1991 experiment, this
study also seeks to extend the original experiment to factors beyond
immorality and responsibility. The omission bias literature discussed
three key factors in the omission bias effect: causality, intentionality,
and regret.
1.3. Replication and procedural adjustments
The steps taken in these and other related studies have not thoroughly addressed all concerns about the generalizability of an omission
1.4.1. Causality
Actions are perceived as more causal than inactions (Baron & Ritov,
2004; Bostyn & Roets, 2016; DeScioli, Bruening, & Kurzban, 2011;
Royzman & Baron, 2002), and causality is key in determining immorality and responsibility. Therefore, harmful commissions would be
2
Google Scholar search of “(omission OR commission OR action OR inaction)” in papers citing Spranca et al. (1991).
2
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
seeing as more causal than harmful omissions, and therefore more
immoral and responsible. In the original study causality was only indirectly assessed using an open-ended question to show that causality
perceptions affected judgment of immorality. We instead assessed
causality perceptions using a clear quantitative scale to directly examine the role of causality in the omission bias.
Participants were first presented with The Tennis Tournament scenario
and then The Eyewitness scenario.
The Tennis Tournament was presented as follows:
John West plays tennis at the Wyncote Tennis Club two or three
times a week. John is the best player belonging to the club, but he is
not good enough to play professionally. The Club holds an annual
tennis tournament, which occasionally attracts a big-name tennis
player in the need of a warm-up before Wimbledon. The first prize is
$20,000, and the prize for the runner-up (who plays in the final but
loses it) is $10,000. This year, Ivan Lendl agreed to play in the
tournament. John and Ivan quickly advanced until they were to
meet in the final. John would of course love to win, but he realizes
that he is at a large disadvantage. The tradition of Wyncote is for
both finalists to meet for dinner at the club before the final the next
morning. While getting dressed for dinner John remembers reading
that Ivan is allergic to Cayenne pepper. He also recalls that house
dressing served in the dining room contains Cayenne pepper. John
thinks, ‘If Ivan eats the house dressing, he will probably get stomach
ache that will keep him up much of the night. Then I’ll have a chance
to win.’After the dinner, Ivan orders first. After he orders his main
course, the waiter asks him whether he prefers the house dressing or
Italian dressing. Ivan does not think that the house dressing might
contain Cayenne pepper.
1.4.2. Intentionality
Actions are perceived as more intentional than inactions (Hayashi,
2015; Kordes-de Vaal, 1996), and intent affects perceptions of morality
and responsibility. In the original study and much of the subsequent
work on omission bias, intentionality was treated as a confound to be
controlled for or fixed (e.g. Royzman & Baron, 2002). We added a
measure of intent to examine the extent to which perceptions of intentionality, and the role in omission bias.
1.4.3. Regret
Historically, the omission bias followed on an earlier demonstration
of action-inaction asymmetries by Kahneman and Tversky (1982)
coined the action-effect. In research on the action-effect, negative
outcomes in everyday life situations are perceived as involving higher
regret if they were a result of action compared to inaction. The actioneffect is considered one of the strongest most replicated findings in the
regret literature (Gilovich & Medvec, 1995). The action-effect was focused on regret and associated counterfactual thinking (thoughts of
what might have been), and the omission bias extended that to show
action-inaction asymmetries in moral situations and decisions made
when faced with the choice between action and inaction (Anderson,
2003; Feldman, Kutscher, & Yay, 2018; Zeelenberg, Van den Bos, Van
Dijk, & Pieters, 2002). The prevalent assumption is that the omission
bias is aligned with action-effect (e.g., DeScioli, Christner, & Kurzban,
2011; Ritov & Baron, 1990), in that everyday life situations and moral
situations would demonstrate a similar action-inaction regret asymmetry (e.g., Baron & Ritov, 2004; Ritov & Baron, 1995; Spranca et al.,
1991). We therefore added a measure of regret to examine whether the
action-effect regret action-inaction asymmetry phenomenon would be
observed in classic omission bias moral scenarios.
In summary, we expected that in addition to Spranca et al.’s original
predictions on perceived immorality and responsibility, participants
would rate harmful outcomes through commission as being more casual, intentional, immoral, accountable, and regretful than through
omissions.
The scenario then ended with one of six outcomes with a manipulation of both commission-omission and intended harmful outcome.
[Commission before choice condition: Before Ivan makes a choice,
John recommends that Ivan try the house dressing. Ivan orders it
and gets a stomach ache, as predicted. If Ivan had said nothing, Ivan
would have ordered Italian dressing, but John does not know this for
sure. John wins the match. Omission condition: Ivan orders the
house dressing and gets a stomach ache, as predicted. John says
nothing. John realized that if he had warned Ivan about the
Cayenne, even after Ivan announced his choice, Ivan would have
ordered Italian dressing. John wins the match. Commission after
choice condition: Ivan orders Italian dressing. John then recommends that Ivan try the house dressing. Ivan changes his mind,
orders the house dressing, and gets stomach ache, as predicted. John
wins the match.]
In three additional endings, the scenarios were exactly as above
with the only change being that Ivan won the match. So despite John’s
behavior of causing Ivan’s allergic reaction, the behavior did not result
in the intended harmful outcome of Ivan losing the match.
The text of the second scenario, The Eyewitness, was presented as
follows:
2. Experiment
2.1. Effect size and power analysis
Peter, a resident of Ohio, is driving through a small town in South
Carolina. At a 4-way stop, he gets into a small accident with a town
resident named Lyle. The accident came about like this: Traveling
north, Lyle approached the 4-way stop and failed either to slow
down or stop. Meanwhile, Peter had just finished stopping and
began to move east through the intersection. Peter noticed that a
car, Lyle’s, was crossing the intersection after having failed to stop.
Peter slammed on his brakes, but too late to prevent his car from
hitting Lyle’s car as it passed in front of him. The accident was
clearly Lyle’s fault because the accident was caused by his failure to
stop. However, because the accident’s cause is not clear from its
effects, the police may believe that Peter failed to stop and that
caused Peter to run into Lyle’s car broadside. Immediately after the
accident, both men exclaimed that is was the other’s fault. When the
police came, Peter told them that the accident was caused by Lyle’s
failure to stop. Lyle told the police that the accident was caused by
Peter’s failure to stop. Unknown to either man, there was an eyewitness to the accident, Ellen. Like Lyle, Ellen is a town resident. She
thought to herself, ‘I know the accident is Lyle’s fault, but I know
Lyle and do not wish him to be punished. The only way that Lyle
We pre-registered a power analysis of the results described in
Spranca et al. (1991) and the analysis is provided in the supplementary
materials (α = 0.05, one tailed, power = 0.95; G*Power 3.1). Unfortunately, the original study did not report means and standard deviations to allow a more accurate estimate of the effect, and we
therefore based our estimates on a conversion of the effect to Cohen’s d
as reported in Table 4 (also see supplementary).
2.2. Participants and procedure
A total of 313 American Amazon Mechanical Turk participants
(Mage = 36.37, SDage = 11.91; 157 females) were recruited online
survey using Turkprime.com (Litman, Robinson, & Abberbock, 2017).
Participants indicated their consent and were presented with two scenarios describing an actor in a position to harm a victim. Participants
were randomly assigned to one condition in each of the two scenarios.
In each scenario, an actor sought to cause a harmful outcome through
either commission or omission and their behavior led either to the intended harmful outcome or to a no-harm outcome for the victim.
3
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
Eyewitness scenario: “Ellen intended to harm Peter.” (1 – Strongly disagree; 7 – Strongly agree).
will be faulted by the police is if I testify that the accident is indeed
Lyle’s fault.’
The scenario then ended with one of four outcomes that created a
manipulation in which Ellen behaved to harm Peter by either commission or omission and then, as an outcome, Peter was either harmed
by being charged or not harmed by not being charged for the accident.
[Commission – harm condition: Ellen told the police that the accident was caused by Peter’s failure to stop. Peter is charged with
failure to stop and fined. Commission – no-harm condition: Ellen told
the police that the accident was caused by Peter’s failure to stop.
Lyle is charged with failure to stop and fined. Omission – harm
condition: Ellen told the police nothing. Peter is charged with failure
to stop and fined. Omission – no-harm condition: Ellen told the
police nothing. Lyle is charged with failure to stop and fined.]
2.3.5. Regret
Participants rated perceived feelings of regret of the actor by indicating their agreement with the following statements – Tennis
Tournament scenario: “John regrets his behavior.”; Eyewitness scenario: “Ellen regrets her behavior.” (1 – Strongly disagree; 7 – Strongly
agree).
3. Results
Descriptives and correlations of the dependent measures are provided in Table 1. Sample size, means, and standard deviations of all
experimental conditions are presented in Table 2.
To test our results, we first ran an independent t-test for each hypothesized condition on each outcome variable. As in the original
study, we present the main findings of our t-tests with a pooled
grouping of the commission before and commission after groups. As a
supplementary analysis, we also tested the interaction between commission and harm using two-way analysis of variance (ANOVA) with
commission versus omission and intended-harm versus no intendedharm, and the contrasts. The findings of both the t-test and ANOVA tests
are presented in Table 3.
Participants were randomly assigned to one of the conditions in
each of the scenarios in a between-subject design. Each condition was
followed by five comprehension questions participants had to answer
correctly to proceed to answering the dependent measures, meant to
ensure participants’ accurate comprehension of the scenarios (see supplementary).
2.3. Measures
2.3.1. Immorality
Following from the original measure used by Spranca et al. (1991),
participants were asked to rate the actor’s morality of the decision in the
given situation on a scale from −100 (as immoral as possible to be in the
situation) to 0 (not immoral at all). This value was then reversed to a
positive integer ranging from 0 (not immoral at all) to 100 (as immoral as
possible to be in the situation).
3.1. Replication results: morality and responsibility
We tested the effects of commission and intended-harm on perceptions of morality and responsibility using a between-person t-test for
main effects and ANOVA for the interaction effects as reported in
Table 3. Unlike the original within-person design, our between-person
design returned far few instances of matched values between conditions
(i.e., in the original study, participants are reported as giving either
higher, lower, or the same attribution of morality or responsibility
between condition, with very many conditions being evaluated as
having the same morality or responsibility). For this reason, our dependent variables across these conditions can be evaluated as continuous rather than ordinal.
Before and after commission conditions were grouped and tested
together as a pooled commission condition for hypothesized relationships. The findings in The Tennis Tournament scenario were in support
of commission-omission asymmetry in immorality attributions (pooled
commission vs. omission: d = −0.45). The effects for intended harm
contrast were weaker (d = 0.11) with Cohen’s d CI including 0, and
with no support for an interaction between the two factors
(np2 = 0.00).
The findings for responsibility judgments were similar to the morality findings. Findings were again in support of commission-omission
asymmetry for responsibility judgments (pooled commission vs. omission: d = −0.53). The effects for intended harm contrast were again
weaker (d = −0.20) with Cohen’s d CI including 0, and with no support
for an interaction between the two factors (np2 = 0.00).
The findings in The Eyewitness scenario regarding perceived immorality were very similar for the main omission bias hypotheses
contrasting commission and omission (d = −0.47). Unlike The Tennis
Tournament scenario, there was also a comparable effect for the harm/
no-harm contrasts with immorality (d = −0.40), with an unexpected
weak interaction (F = 4.01, p = .046, np2 = 0.01). The same pattern
emerged for responsibility judgments having very similar effects to the
immorality ratings (commission vs. omission: d = −0.40; harm vs. noharm: d = −0.49; interaction: np2 = 0.002).
We concluded a successful replication of the omission bias omissioncommission asymmetry reported in the original study. The findings
2.3.2. Responsibility
Following from the original measure used by Spranca et al. (1991),
participants rated perceived moral responsibility by putting a dollar
value on the penalty that the actor should have imposed on them if they
were caught in their potentially harmful behavior. In the first scenario,
participants were presented with the following: “Suppose that Ivan
found out that John knew about the dressing and Ivan’s allergy and Ivan
is now suing John. You are on the jury and are convinced by the evidence that the case is exactly as described above”. In the second scenario, participants were presented with the following: “Suppose that
Peter found out that Ellen told the police that it was Peter’s failure to
stop. You are on the jury and are convinced by the evidence that the
case is exactly as described above”. In both scenarios, participants were
asked to provide a numerical answer with an unrestricted range with
the following: “How much money, if any, do you think [Ivan / Peter]
should receive in compensation? (in USD)”. Responsibility compensation ratings were transformed using the natural log to address skewness.
2.3.3. Causality
Building on the original study’s open-ended measure of causality,
we added a quantitative measure of causality. Participants rated perceived causality by indicating their agreement to the following statements – Tennis Tournament scenario: “John understood that his behavior would affect Ivan in the way that it did”; Eyewitness scenario:
“Ellen understood that her behavior would affect Peter.” (1 – Strongly
disagree; 7 – Strongly agree).
2.3.4. Intentionality
Participants rated perceived intentionality by indicating their
agreement to the following statements – Tennis Tournament scenario:
“John intended for his behavior to affect Ivan in the way that it did.”;
4
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
Table 1
Means, standard deviations, and correlations for all variables under all conditions.
M
SD
Scenario 1: The Tennis Tournament
1. Immoralitya
74.70
24.77
2. Responsibilityb
6.21
4.03
3. Causalityc
6.13
1.27
6.11
1.35
2.60
1.43
Scenario 2: The Eyewitness
1. Immoralitya
71.24
26.48
4. Intentionality
5. Regret
c
c
b
2. Responsibility
6.35
3.82
3. Causality
6.16
1.04
4. Intentionalityc
4.18
1.73
5. Regretc
2.95
1.49
c
Commission
Harm outcome
1
2
3
4
0.21**
[0.10, 0.31]
0.24**
[0.14, 0.35]
0.12*
[0.01, 0.23]
0.13*
[0.02, 0.24]
−0.09
[−0.20, 0.02]
−0.06
[−0.17, 0.06]
0.10
[−0.01, 0.21]
0.15**
[0.04, 0.26]
0.21**
[0.10, 0.31]
−0.14*
[−0.25, −0.03]
0.20**
[0.09, 0.30]
0.33**
[0.23, 0.42]
0.27**
[0.16, 0.37]
−0.24**
[−0.34, −0.13]
0.09
[−0.02, 0.20]
0.10
[−0.01, 0.21]
−0.11
[−0.21, 0.01]
0.77**
[0.72, 0.81]
−0.26**
[−0.36, −0.15]
−0.25**
[−0.35, −0.15]
0.23**
[0.12, 0.33]
0.20**
[0.09, 0.30]
0.10
[−0.01, 0.21]
0.28**
[0.17, 0.38]
−0.13*
[−0.24, −0.02]
0.20**
[0.09, 0.30]
0.24**
[0.13, 0.34]
−0.02
[−0.13, 0.09]
0.08
[−0.03, 0.19]
0.05
[−0.06, 0.16]
0.23**
[0.12, 0.33]
0.40**
[0.30, 0.49]
0.16**
[0.05, 0.27]
−0.25**
[−0.36, −0.15]
0.16**
[0.05, 0.26]
0.13*
[0.02, 0.24]
−0.09
[−0.19, 0.03]
0.22**
[0.12, 0.33]
−0.33**
[−0.42, −0.22]
−0.24**
[−0.34, −0.13]
N = 313. *p < .05, **p < .01, ***p < .001. Values in square brackets indicate the 95% confidence interval for each correlation (Cumming, 2014).
a
Omission condition for The Tennis Tournament scenario is reported using the pooled condition of commission before and commission after. Immorality scale is
from 0 to 100.
b
Responsibility is a positive number with no range restriction, and was log-transformed to address skewness.
c
Causality, intentionality, and regret are scale of 1–7.
regarding intended harm contrasts were not as clear, and they correspond to the mixed findings in the original study (only 8 and 6 of the 57
subjects rating higher immorality for the harmful outcome in The
Tennis Tournament and The Eyewitness scenarios respectively, with
one subject showing the opposite effect). Based on the null to very weak
interactions, we conclude that it is most likely that the two factors do
not interact.
generalizability and reliability of early studies. A summary and comparison of findings in the original and replication are provided in
Table 4. We conclude a successful replication of the baseline omission
bias with slight deviations from the original findings regarding outcome
bias. We found effects for the added extensions, with medium effects for
intentionality and weaker effects for causality and regret, with the regret effects opposite from the literature on the classic action-effect
(Kahneman & Tversky, 1982).
3.2. Extension results: causality, intention, and regret
4.1. Immorality and responsibility: support for an omission bias
Examining omission-commission contrasts, effects were generally
consistent across the two scenarios for causality (Tennis Tournament:
d = −0.27; Eyewitness: d = −0.21), intentionality (Tennis
Tournament: d = −0.28; Eyewitness: d = −0.58), and regret (Tennis
Tournament: d = 0.18; Eyewitness: d = 0.26). Also consistent, was the
very weak interaction effects between omission-commission and harm/
no-harm in both scenarios for these factors (np2 < 0.005). The findings
regarding the intended harm contrasts, however, were not consistent
across the two scenarios. Effects were stronger and significant for
Tennis Tournament scenario (−0.43 < d < 0.29, p < .011) compared to the weaker and non-significant effects in The Eyewitness
scenario (−0.17 < d < 0.04, p > .137).
The correlations between the factors were consistent across the two
scenarios. Higher immorality ratings were associated with higher ratings for causality (r = 0.33 to 0.40) and intentionality (r = 0.27 to
0.16), and lower ratings of regret (r = −0.24 to −0.25). The stronger
the perceived connection between a person and the outcome (causality)
and perceived intent to harm, the less likely the actor is perceived to be
regretful of the act, the less moral the act seems, and the higher the
compensation that was awarded to the target. Initial mediation analyses
examining the role of causality and intentionality are provided in the
supplementary.
Based on the original study, we expected participants to rate actions
leading to harm through omission as less immoral than corresponding
harm through commission. Our findings are in support of these omission bias findings, with a consistent effect in both scenarios of Cohen’s d
of −0.45 to −0.47, slightly weaker than expected, yet reasonable given
the conversions and modifications in analyses.
4.2. Causality and intentionality: support for classic omission bias findings
We extended the original study by Spranca et al. (1991) by adding
possible factors associated with the omission-commission effect, causality and intent, to compliment morality and responsibility. Compared
to commissions, omissions were generally perceived as less causal
(d = −0.21 to −0.27) and less intentional (d = −0.28 to −0.58).
These are in line with the idea that omissions are perceived as nondecisions involving less deliberation and intent (Hayashi, 2015; Kordesde Vaal, 1996; Ritov & Baron, 1992).
Going beyond the experimental design, the findings from the correlational analyses were in line with the literature on the omission bias.
The stronger the perceived connection between a person and the outcome (causality) and perceived intent to harm, the more immoral the
act seems, and the higher the compensation that was awarded to the
target. These findings help to support the model proposed in the early
work on omission bias which theorized that perceptions of increased
causality and intentionality of commission over omission drove the
bias.
4. Discussion
We set out to replicate and extend the classic Spranca et al. (1991)
demonstration of the omission bias to address concerns about the
5
6
82.36
7.59
6.50
6.62
2.26
SD
b
80.96
7.35
5.94
5.83
2.70
19.93
3.30
1.61
1.66
1.34
76.93
5.76
6.06
6.00
2.63
M
79.75
7.85
6.17
4.75
2.80
21.88
3.40
1.08
1.73
1.65
73.37
6.71
6.10
3.92
3.24
24.24
3.59
1.26
1.48
1.54
74.99
6.38
6.35
4.59
2.71
M
M
SD
27.57
4.38
1.25
1.08
1.25
SD
M
64.32
5.32
5.98
6.06
2.46
M
Commission
(n = 79)
26.18
3.92
0.98
0.91
1.51
SD
Omission
(n = 79)
77.69
7.28
6.50
6.56
2.36
M
Commission
(n = 76)
28.44
3.97
0.86
0.86
1.36
SD
No-harm outcome
73.12
6.96
6.49
6.51
2.45
M
27.15
3.58
0.79
1.70
1.28
SD
21.49
3.68
1.19
1.41
1.39
SD
Commission before
(n = 54)
Harm outcome
23.01
3.88
1.09
0.97
1.66
SD
M
M
SD
Commission after
(n = 54)
Omission
(n = 50)
Commission before
(n = 51)
Commission after
(n = 50)
Pooled commission
(n = 101)
No-harm outcome
Harm outcome
Immorality scale is from 0 to 100.
Responsibility is a positive number with no range restriction and was log-transformed to address skewness.
c
Causality, intentionality, and regret are scale of 1–7.
a
Immoralitya
Responsibilityb
Causalityc
Intentionalityc
Regretc
Scenario 2:
The Eyewitness
Immoralitya
Responsibilityb
Causalityc
Intentionalityc
Regretc
Scenario 1:
The Tennis
Tournament
Table 2
Means and standard deviations for all conditions in The Tennis Tournament scenario and The Eyewitness scenario.
78.94
6.55
6.00
5.92
2.67
M
57.19
4.50
6.00
3.48
3.04
M
Omission
(n = 79)
20.73
3.57
1.40
1.54
1.36
SD
Pooled commission
(n = 108)
70.20
4.35
5.83
5.67
3.07
M
Omission
(n = 54)
27.01
3.96
0.97
1.73
1.44
SD
23.95
4.07
1.37
1.64
1.46
SD
J. Jamison, et al.
Journal of Experimental Social Psychology 89 (2020) 103977
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
Table 3
Main-effects and interactions.
Scenario 1:
The Tennis Tournament
Omission vs commission (before)
Omission vs commission (after)
Commission (before) vs commission (after)
t
p
d
[95% CI]
t
p
d
[95% CI]
t
p
d
[95% CI]
Immorality
Responsibility
Causality
Intentionality
Regret
−2.19
−2.73
−2.20
−2.17
1.23
.030
.007
.029
.031
.219
−0.30
−0.38
−0.30
−0.30
0.17
[−0.58, −0.03]
[−0.65, −0.10]
[−0.58, −0.03]
[−0.57, −0.03]
[−0.10, 0.44]
−4.34
−4.88
−1.63
−1.82
1.43
< .001
< .001
.104
.071
.154
−0.60
−0.68
−0.23
−0.25
0.20
[−0.88, −0.32]
[−0.96, −0.39]
[−0.50, 0.05]
[−0.53, 0.02]
[−0.08, 0.47]
2.04
2.18
−0.32
−0.20
−0.26
.043
.030
.749
.843
.793
0.28
0.30
−0.04
−0.03
−0.04
[0.01, 0.55]
[0.03, 0.58]
[−0.32, 0.23]
[−0.3, 0.24]
[−0.31, 0.23]
Scenario 1:
The Tennis Tournament
Omission vs commission (pooled)
No-harm outcome vs harm outcome
Omission × no-harm outcome
t
p
d
[95% CI]
t
p
d
[95% CI]
F
p
ηp2
[95% CI]
Immorality
Responsibility
Causality
Intentionality
Regret
−3.65
−4.27
−2.17
−2.27
1.55
< .001
< .001
.031
.024
.123
−0.45
−0.53
−0.27
−0.28
0.18
[−0.69, −0.21]
[−0.78, −0.29]
[−0.50, −0.03]
[−0.52, −0.04]
[−0.05, 0.42]
0.98
−1.77
−2.67
−3.83
2.57
.328
.078
.008
< .001
.011
0.11
−0.20
−0.30
−0.43
0.29
[−0.11, 0.33]
[−0.42, 0.02]
[−0.53, −0.08]
[−0.65, −0.20]
[0.07, 0.51]
0.63
0.07
1.34
0.65
0.80
.428
.796
.247
.421
.371
0.00
0.00
0.00
0.00
0.00
[0.00, 0.02]
[0.00, 0.01]
[0.00, 0.03]
[0.00, 0.02]
[0.00, 0.03]
Scenario 2:
The Eyewitness
Omission vs commission
t
p
d
Immorality
Responsibility
Causality
Intentionality
Regret
−4.13
−3.53
−1.83
−5.14
2.30
< .001
< .001
.068
< .001
.022
−0.47
−0.40
−0.21
−0.58
0.26
No-harm outcome vs harm outcome
Omission × no-harm outcome
[95% CI]
t
p
d
[95% CI]
F
p
ηp2
[95% CI]
[−0.69, −0.24]
[−0.62, −0.17]
[−0.43, 0.02]
[−0.81, −0.35]
[0.04, 0.48]
−3.55
−4.36
0.35
−1.49
−0.90
< .001
< .001
.724
.137
.367
−0.40
−0.49
0.04
−0.17
−0.10
[−0.63, −0.17]
[−0.72, −0.26]
[−0.18, 0.26]
[−0.39, 0.05]
[−0.32, 0.12]
4.01
0.77
1.48
0.59
0.11
.046
.380
.226
.444
.746
0.01
0.00
0.00
0.00
0.00
[0.00, 0.05]
[0.00, 0.03]
[0.00, 0.03]
[0.00, 0.02]
[0.00, 0.02]
Note. Bolded values indicate consistent patterns of significant findings (p < .05) across the two scenarios. Italicized values mark inconsistent significant findings
(p < .05). Confidence intervals are reported at 95%. Omission condition for The Tennis Tournament scenario is reported using the pooled condition of commission
before and commission after.
outcome that deviates from expectation. We hypothesized that participants would perceive actors as being more regretful for taking action
that would immorally harm another person rather than allowing that
harm through inaction. Yet it is plausible that participants were focused
on the regret that actors would feel for not taking more direct action
towards their goal of personal or interpersonal gain.
Another possible explanation for the regret finding is the side-taking
hypothesis (De Freitas and Johnson, 2018; DeScioli, 2016; DeScioli &
Kurzban, 2013). This states that group members side against a wrongdoer who has performed an action that is perceived morally wrong by
also attributing lack of remorse or regret. The negative relationship
observed between the positive characteristic of regret and the negative
characteristics of immorality, causality, and intentionality is in support
of this explanation. Future research may be able to explore the true
mechanisms of regret in such scenarios.
The role of this study's between-subject design is also worth noting
regarding intentionality. The intended purpose of the within-subject
design in the original study and other follow-ups by those authors (e.g.,
Baron & Ritov, 2009; Royzman & Baron, 2002) was to hold intentionality constant in the mind of the subject. Using a within-subject
design was meant to make clear to participants that the intention of the
actor was already formed when the decision presented itself. We adjusted to a between-subject design to test the generalizability of these
earlier findings to a design that is closer to everyday life situations,
where often information about behaviors and outcomes is incomplete.
It also allowed us to measure and analyze variations in judgements of
intentionality across scenarios. For more discussion on this, supplemental analyses of the relationship between intentionality and omission
behaviors are reported in the supplementary. Our results provide empirical support for a role of intentionality in judgements of omission
behaviors that may be fertile for future study.
4.4. Outcome bias: deviations from original findings
4.3. Regret: deviation from the action-effect
Spranca et al. (1991) also examined an “outcome bias”, comparing
participants' judgements of an actor's intentionally harmful behavior
that lead either to the intended harmful outcome or to a no-harm
outcome for the victim. The replication of these comparisons yielded
inconsistent findings across the two scenarios. In The Eyewitness scenario, the strongest effects were for immorality and responsibility
(d = −0.40 and −0.49, respectively), with weak effects for causality,
intentionality, and regret (d = 0.04, −0.17, and −0.10, respectively),
whereas in The Tennis Tournament scenario, the strongest effects were
for causality, intention, and regret (d = −0.30, −0.42, and 0.29, respectively), with weaker effects for morality and responsibility
(d = −0.11 and −0.20, respectively).
These suggest some underlying difference between the two scenarios regarding the harmful outcome, despite the consistent findings
for the omission bias action-inaction asymmetry. Spranca et al. (1991)
noted that outcome bias could be related to the within-subject design,
The classic action-effect (Kahneman & Tversky, 1982) findings were
that actions leading to a negative outcome are regretted more than
inactions leading to the same negative outcomes. We added a regret
measure to examine whether the action-effect findings would extend to
situations of morality involving intended harmful behavior. Our findings were opposite to the expected action-effect omission-commission
asymmetry with participants rating omissions as more regretted than
commissions (d = 0.18 to 0.26).
One explanation for this surprising finding may be an intermingling
of the perception of an actors' regret for their behavior with their regret
for the outcome. In typical action-effect scenarios, actors behave in a
way that is morally neutral but are faced with an outcome that deviates
from expectations, such as losing money over an investment. In this
study's omission bias scenarios, the actors behaved immorally to harm
others for personal or interpersonal gain, and then are faced with an
7
Journal of Experimental Social Psychology 89 (2020) 103977
Note. Bolded values indicate consistent patterns of significant findings (p < .05) across the two scenarios in the replication and extension study. Italicized values mark inconsistent significant findings (p < .05) across
the two scenarios in the replication and extension study. Confidence intervals are reported at 95%. Results of original study calculated here using one sample T-Tests and are reported as Cohen's d. Results of the
replication study calculated using independent sample T-Tests and effects are reported as Cohen's d. Replication outcomes taken from LeBel, Vanpaemel, Cheung, & Campbell, 2019.
[−0.72, −0.26]
[−0.18, 0.26]
[−0.39, 0.05]
[−0.32, 0.12]
Responsibility
Causality
Intentionality
Regret
Regret
Immorality
No-harm outcomes will be associated with a bias
towards lower attributions of
Causality
Intentionality
−0.40 [−0.67,
−0.13]
−0.34 [−0.61,
−0.07]
0.18
0.11
−0.49
0.04
−0.17
−0.10
Inconsistent
No-signal
0.26
−0.40
−0.21
−0.58
−0.27
−0.28
[−0.42, 0.02]
[−0.53, −0.08]
[−0.65, −0.20]
[0.07, 0.51]
Consistent
Signal
−0.40
−0.53
−1.21 [−1.55,
−0.86]
−1.26 [−1.61,
−0.90]
Immorality
Omission will be associated with a bias towards lower
attributions of
Responsibility
−0.20
−0.30
−0.43
0.29
Consistent
Signal
[−0.69,
−0.24]
[−0.62,
−0.17]
[−0.43, 0.02]
[−0.81,
−0.35]
[0.04, 0.48]
[−0.63, −0.17]
−0.47
[−0.69,
−0.21]
[−0.78,
−0.29]
[−0.50, −0.03]
[−0.52,
−0.04]
[−0.05, 0.42]
[−0.11, 0.33]
−0.45
Scenario 1:
The Tennis
Tournament
Rater attributions to actor
characteristics
Experimental conditions
−1.46 [−1.83,
−1.08]
Consistent/
inconsistent
Signal/no
signal
Scenario 2:
The Eyewitness
Scenario 1:
The Tennis Tournament
Results of original study
Hypotheses
Table 4
Summary of replication and extension results.
Scenario 2:
The Eyewitness
Results of replication and extension study
Replication outcome
J. Jamison, et al.
explaining that “we placed the cases next to each other to determine
whether anyone knowingly evaluates decisions according to their outcome” (p. 82; italics in original text). Since the experimental design was
adapted to a between-subject design, the subjects in the replication
were not aware of the outcome differences between conditions, especially in the first scenario where participants had not yet been exposed
to any other outcomes.
A possible distinction between the two scenarios involves the reason
for harm. In The Tennis Tournament scenario harm is inflicted for
personal benefit (winning the tournament), whereas in The Eyewitness
scenario harm is inflicted to help a member of the ingroup (protect a
fellow resident). Factors related to causality, intention, and regret could
be more relevant when situation is more complex in terms of moral
reasoning in an interpersonal context. Future research on outcome bias
should examine what in these two scenarios may have led to such
differences.
4.5. Limitations and future directions
We set out to replicate a cornerstone omission bias study following
growing concerns in psychological science regarding the reproducibility, replicability, reliability, and generalizability of classic psychological effects. Our replication updates the original study to meet current open-science standards including a pre-registered study design,
sample size determined by power analysis, advanced statistical analyses, robust results reporting with effect-size estimations, and openly
available materials, data, and code. However, in conducting a direct
close replication (according to the criteria set by LeBel et al. (2018) we
decided on several important theoretical and empirical adjustments to
go try and go beyond the original to add extensions that would shed
new insights on the phenomenon. We adjusted the study design to a
between-subject design to test the generalizability of the effect and
added extensions with hypotheses for causality, intentionality, and regret. The close replication and extension approach adds to the literature
and more recent theorical developments (e.g. prosocial motivations in
Levine et al., 2018 and counterfactual thinking in Henne, Niemi,
Pinillos, De Brigard, & Knobe, 2019).
Our findings are in line with several other recent empirical demonstrations finding support for omission bias using adjusted conceptual replications yet recognizing boundary conditions (Bostyn &
Roets, 2016; Siegel, Crockett, & Dolan, 2017). An important future direction would be to conduct a systematic review and meta-analysis of
the omission bias literature (Yeung et al., 2020).
In any close replication, replicators face multiple decisions on tradeoffs in design. In this study, a methodological limitation worth noting
was our decision to administer two randomized conditions in succession to each participant. Consequently, Scenario 1 offers the clearest
picture of between-subjects results, with responses to Scenario 2 possibly affected in some way from exposure to Scenario 1 However, we do
not believe these affect results for a number of reasons. First, we successfully replicated the core results from the original findings. Second,
there were major differences in nature of the scenarios (personal, interpersonal) as well as their designs and complexity (Tennis
Tournament scenario with 3 × 2 structure versus Eyewitness scenario
with 2 × 2 design) minimize such concerns. Third, other similar replication work directly tested order effects of two different experiments
testing the same phenomenon concluded no order effects (Kutscher &
Feldman, 2019; Ziano, Yao, Gao, & Feldman, 2020). Future studies
would ideally address this limitation by fixing participants to one
condition throughout both scenarios, randomizing order, or directly
testing order effects.
Another decision we faced was about study design, whether to run
the same within-person design or switch to a between-person design.
We saw value in adjusting the design to a between design, given our
reading of the literature and the debate about within-between designs,
and the extensions we planned to examine causality and intentionality.
8
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
Further, we sought to make the study more realistic and representative
of real-life situations, and learned from previous experience involving
challenges when using within designs in our specific target online
sample. Yet, that decision also has limitations. Differences between
conditions become salient in the within-subject design used in the
original study, making circumstances and outcomes clearer to readers,
and this may have affected our results regarding outcome bias. Future
research may further contrast the two designs against each other in a
single study to examine whether this truly has any effect on either
outcome bias or omission bias.
Baron, J., & Ritov, I. (2009). The role of probability of detection in judgements of punishment. Journal of Legal Analysis, 1, 553–590. https://doi.org/10.1093/jla/1.2.553.
Bostyn, D. H., & Roets, A. (2016). The morality of action: The asymmetry between
judgments of praise and blame in the action–omission effect. Journal of Experimental
Social Psychology, 63, 19–25. https://doi.org/10.1016/j.jesp.2015.11.005.
Connolly, T., & Reb, J. (2003). Omission bias in vaccination decisions: where's the
"omission"? Where's the "bias"? Organizational Behavior and Human Decision Processes,
91, 186–202. https://doi.org/10.1016/S0749-5978(03)00057-8.
Connolly, T., & Reb, J. (2012a). Regret aversion in reason-based choice. Theory and
Decision, 73(1), 35–51. https://doi.org/10.1007/s11238-011-9269-0.
Connolly, T., & Reb, J. (2012b). Toward interactive, Internet-based decision aid for
vaccination decisions: Better information alone is not enough. Vaccine, 30(25),
3813–3818. https://doi.org/10.1016/j.vaccine.2011.12.094.
Cumming, G. (2014). The new statistics: Why and how. Psychological Science, 25(1), 7–29.
Cushman, F., & Young, L. (2011). Patterns of moral judgment derive from nonmoral
psychological representations. Cognitive Science, 35, 1052–1075. https://doi.org/10.
1111/j.1551-6709.2010.01167.x.
De Freitas, J., & Johnson, S. G. (2018). Optimality bias in moral judgment. Journal of
Experimental Social Psychology, 79, 149–163. https://doi.org/10.1016/j.jesp.2018.07.
011.
DeScioli, P. (2016). The side-taking hypothesis for moral judgment. Current Opinion in
Psychology, 7, 23–27. https://doi.org/10.1016/j.copsyc.2015.07.002.
DeScioli, P., Bruening, R., & Kurzban, R. (2011). The omission effect in moral cognition:
Toward a functional explanation. Evolution and Human Behavior, 32, 204–215.
https://doi.org/10.1016/j.evolhumbehav.2011.01.003.
DeScioli, P., Christner, J., & Kurzban, R. (2011). The omission strategy. Psychological
Science, 22, 442–446. https://doi.org/10.1177/0956797611400616.
DeScioli, P., & Kurzban, R. (2013). A solution to the mysteries of morality. Psychological
Bulletin, 139, 477. https://doi.org/10.1037/a0029065.
Feldman, G., Kutscher, L., & Yay, T. (2018). What is action, what is inaction? A review of
action-inaction biases and recommendations for term use and typology. Retrieved
July 2018 https://www.researchgate.net/publication/320409218_What_is_action_
what_is_inaction_A_review_of_action-inaction_biases_and_recommendations_for_term_
use_and_typology.
Gilovich, T., & Medvec, V. H. (1995). The experience of regret: What, when, and why.
Psychological Review, 102, 379. https://doi.org/10.1037/0033-295X.102.2.379.
Hayashi, H. (2015). Omission bias and perceived intention in children and adults. British
Journal of Developmental Psychology, 33, 237–251. https://doi.org/10.1111/bjdp.
12082.
Henne, P., Niemi, L., Pinillos, Á., De Brigard, F., & Knobe, J. (2019). A counterfactual
explanation for the action effect in causal judgment. Cognition, 190, 157–164.
https://doi.org/10.1016/j.cognition.2019.05.006.
Hershey, J. C., Asch, D. A., Thumasathit, T., Meszaros, J., & Waters, V. V. (1994). The
roles of altruism, free riding, and bandwagoning in vaccination decisions.
Organizational Behavior and Human Decision Processes, 59(2), 177–187. https://doi.
org/10.1006/obhd.1994.1055.
Kahneman, D., & Tversky, A. (1982). The psychology of preferences. https://doi.org/10.
1038/scientificamerican0182-160.
Klein, R. A., Vianello, M., Hasselman, F., Adams, B. G., Adams, R. B., Jr., Alper, S., &
Batra, R. (2018). Many Labs 2: Investigating variation in replicability across samples
and settings. Advances in Methods and Practices in Psychological Science, 1(4), 443–490.
https://doi.org/10.1177/2515245918810225.
Kordes-de Vaal, J. H. (1996). Intention and the omission bias: Omissions perceived as
nondecisions. Acta Psychologica, 93, 161–172. https://doi.org/10.1016/00016918(96)00027-3.
Kutscher, L., & Feldman, G. (2019). The impact of past behavior normality on regret:
Replication and extension of three experiments of the exceptionality effect. Cognition
and Emotion, 33(5), 901–914.
LeBel, E. P., McCarthy, R. J., Earp, B. D., Elson, M., & Vanpaemel, W. (2018). A unified
framework to quantify the credibility of scientific findings. Advances in Methods and
Practices in Psychological Science, 1, 389–402. https://doi.org/10.1177/
2515245918787489.
LeBel, E. P., Vanpaemel, K. U. L., Cheung, I., & Campbell, L. (2019). A brief guide to
evaluate replications. Meta-Psychology, 3, 1–9. https://doi.org/10.15626/MP.2018.
843.
Levine, E., Hart, J., Moore, K., Rubin, E., Yadav, K., & Halpern, S. (2018). The surprising
costs of silence: Asymmetric preferences for prosocial lies of commission and omission. Journal of Personality and Social Psychology, 114(1), 29. https://doi.org/10.
1037/pspa0000101.
Litman, L., Robinson, J., & Abberbock, T. (2017). TurkPrime.com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behavior Research
Methods, 49, 433–442. https://doi.org/10.3758/s13428-016-0727-z.
Meszaros, J. R., Asch, D. A., Baron, J., Hershey, J. C., Kunreuther, H., & SchwartzBuzaglo, J. (1996). Cognitive processes and the decisions of some parents to forego
pertussis vaccination for their children. Journal of Clinical Epidemiology, 49, 697–703.
https://doi.org/10.1016/0895-4356(96)00007-8.
N’gbala, A., & Branscombe, N. R. (1997). When does action elicit more regret than inaction and is counterfactual mutation the mediator of this effect? Journal of
Experimental Social Psychology, 33, 324–343. https://doi.org/10.1006/jesp.1996.
1322.
Petrinovitch, L., & O'Neill, P. (1996). Influence of wording and framing effects on moral
intuitions. Ethology and Sociobiology, 17(3), 145–171. https://doi.org/10.1016/01623095(96)00041-6.
Reb, J., & Connolly, T. (2010). The effects of action, normality, and decision carefulness
on anticipated regret: Evidence for a broad mediating role of decision justifiability.
Cognition and Emotion, 24, 1405–1420. https://doi.org/10.1080/
Open practices
We pre-registered the experiment on the Open Science Framework
and data collection was launched later that day. Pre-registration, power
analyses, and all materials used are available in the supplementary
materials. These together with data and code were shared on the Open
Science Framework: https://osf.io/9gsqe/; Pre-registration link:
https://osf.io/6nn57.
Financial disclosure/funding
None.
Declaration of competing interest
The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
Acknowledgments
None.
Authorship declaration
Tijen worked under the supervision of Gilad at Maastricht
University for conducting the pre-registered replication and meta-analysis on omission bias as part of her masters thesis. Tijen wrote the preregistration, with verification and registration by Gilad. Tijen summarized the methods and results and wrote an initial draft. Gilad wrote
the first journal submission draft, with verification of analyses and results, and write-up of the relevant literature and implications.
Following review, John verified and reanalyzed the results, ran additional analyses, and revised the manuscript for submission, addressing
all review comments. Gilad and John finalized manuscript for submission.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.jesp.2020.103977.
References
Anderson, C. J. (2003). The psychology of doing nothing: Forms of decision avoidance
result from reason and emotion. Psychological Bulletin, 129, 139–167. https://doi.org/
10.1037/0033-2909.129.1.139.
Asch, D. A., Baron, J., Hershey, J. C., Kunreuther, H., Meszaros, J., Ritov, I., & Spranca, M.
(1994). Omission bias and pertussis vaccination. Medical Decision Making, 14,
118–123. https://doi.org/10.1177/0272989X9401400204.
Baron, J., & Hershey, J. C. (1988). Outcome bias in decision evaluation. Journal of
Personality & Social Psychology, 54, 569–579. https://doi.org/10.1037/0022-3514.54.
4.569.
Baron, J., & Ritov, I. (1994). Reference points and omission bias. Organizational Behavior
and Human Decision Processes, 59, 475–498. https://doi.org/10.1006/obhd.1994.
1070.
Baron, J., & Ritov, I. (2004). Omission bias, individual differences, and normality.
Organizational Behavior and Human Decision Processes, 94, 74–85. https://doi.org/10.
1016/j.obhdp.2004.03.003.
9
Journal of Experimental Social Psychology 89 (2020) 103977
J. Jamison, et al.
02699930903512168.
Ritov, I., & Baron, J. (1990). Reluctance to vaccinate: Omission bias and ambiguity.
Journal of Behavioral Decision Making, 3, 263–277. https://doi.org/10.1002/bdm.
3960030404.
Ritov, I., & Baron, J. (1992). Status-quo and omission biases. Journal of Risk and
Uncertainty, 5, 49–61. https://doi.org/10.1007/BF00208786.
Ritov, I., & Baron, J. (1995). Outcome knowledge, regret, and omission bias.
Organizational Behavior and Human Decision Processes, 64, 119–127. https://doi.org/
10.1006/obhd.1995.1094.
Royzman, E. B., & Baron, J. (2002). The preference for indirect harm. Social Justice
Research, 15, 165–184. https://doi.org/10.1023/A:1019923923537.
Siegel, J. Z., Crockett, M. J., & Dolan, R. J. (2017). Inferences about moral character
moderate the impact of consequences on blame and praise. Cognition, 167, 201–211.
https://doi.org/10.1016/j.cognition.2017.05.004.
Spranca, M., Minsk, E., & Baron, J. (1991). Omission and commission in judgment and
choice. Journal of Experimental Social Psychology, 27, 76–105. https://doi.org/10.
1016/0022-1031(91)90011-T.
Willemsen, P., & Reuter, K. (2016). Is there really an omission effect? Philosophical
Psychology, 29(8), 1142–1159.
Yeung, S., Yay, T., & Feldman, G. (2020). Omission bias: A meta-analysis. Manuscript in
preparation based on Yay. masters thesis. Retrieved March 2020 from https://osf.io/
vdb3k/?view_only=91512350e10747b0a4767d92339b30ff.
Zeelenberg, M., Van den Bos, K., Van Dijk, E., & Pieters, R. (2002). The inaction effect in
the psychology of regret. Journal of Personality and Social Psychology, 82, 314. https://
doi.org/10.1037/0022-3514.82.3.314.
Ziano, I., Yao, D., Gao, Y., & Feldman, G. (2020). Impact of ownership on liking and value:
Replication of three ownership effect experiments. Journal of Experimental Social
Psychology. Retrieved September 2019 from https://www.researchgate.net/
publication/330423092_Impact_of_ownership_on_liking_and_value_Replications_and_
extensions_of_three_ownership_effect_experiments.
Zwaan, R. A., Etz, A., Lucas, R. E., & Donnellan, M. B. (2018). Making replication
mainstream. Behavioral and Brain Sciences, 41. https://doi.org/10.1017/
S0140525X17001972.
John Jamison is a PhD student with the management department at the Hong Kong
University of Science and Technology.
Tijen Yay is a graduate of the Maastricht University work and social psychology department masters program. For her thesis under the guidance of Gilad Feldman she
conducted a pre-registered replication and pre-registered meta-analysis of the omission
bias.
Gilad Feldman is an assistant professor at the Department of Psychology, University of
Hong Kong. His research focuses on judgment decision-making and open-science.
10
CommissionHarm
(binary)
(binary)
CommissionHarm
(categorical)
(categorical)
Immorality Responsibility
Responsibility
(log)
Causation
(raw) Intent
1
1 CommissionHarm
100 10.30899
30000
7
0
0 Omission No harm
0 8.853808
7000
3
1
0 CommissionNo harm
90 8.517393
5000
7
0
1 Omission Harm
28 9.21044
10000
6
1
0 CommissionNo harm
75
0
0
7
0
1 Omission Harm
93 3.931826
50
6
1
1 CommissionHarm
100 11.51294
100000
7
0
0 Omission No harm
29
0
0
6
1
1 CommissionHarm
34 8.517393
5000
3
1
0 CommissionNo harm
51 2.397895
10
7
0
1 Omission Harm
83 9.21044
10000
7
1
0 CommissionNo harm
85 2.397895
10
7
1
1 CommissionHarm
100 11.51294
100000
7
0
0 Omission No harm
50
0
0
5
0
0 Omission No harm
50 9.21044
10000
7
1
0 CommissionNo harm
88 8.517393
5000
6
0
0 Omission No harm
81 6.908755
1000
6
0
1 Omission Harm
75 5.303305
200
6
1
1 CommissionHarm
85 9.21044
10000
7
1
0 CommissionNo harm
0 6.216606
500
7
0
1 Omission Harm
33 6.908755
1000
5
0
0 Omission No harm
50 9.21044
10000
7
1
1 CommissionHarm
81
0
0
6
1
1 CommissionHarm
90 8.517393
5000
7
1
0 CommissionNo harm
0 8.699681
6000
6
1
0 CommissionNo harm
39 3.610918
36
6
0
1 Omission Harm
70 3.931826
50
7
0
0 Omission No harm
50 6.908755
1000
5
1
1 CommissionHarm
63 8.517393
5000
7
0
1 Omission Harm
70 10.12667
25000
7
1
0 CommissionNo harm
81 8.517393
5000
6
0
0 Omission No harm
65 7.601402
2000
7
1
1 CommissionHarm
96 6.908755
1000
7
1
0 CommissionNo harm
58 8.517393
5000
6
0
0 Omission No harm
11
0
0
5
0
1 Omission Harm
76 11.51294
100000
7
1
1 CommissionHarm
100 10.12667
25000
7
1
0 CommissionNo harm
90 10.12667
25000
6
1
1 CommissionHarm
100 6.216606
500
6
0
1 Omission Harm
98 8.517393
5000
7
0
0 Omission No harm
50 6.908755
1000
6
1
1 CommissionHarm
24
0
0
5
0
0 Omission No harm
76 6.216606
500
6
1
0 CommissionNo harm
100
0
0
7
0
1 Omission Harm
31 2.397895
10
4
0
1 Omission Harm
89 6.398595
600
7
7
2
2
5
3
4
7
3
6
3
2
5
7
1
1
6
3
5
6
6
5
1
5
2
2
3
2
1
1
3
5
2
3
4
6
5
7
5
6
3
3
3
6
4
1
2
1
0
1
1
0
1
0
0
0
1
1
1
0
0
0
0
1
1
1
1
1
0
0
1
0
1
0
0
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
0
0
1
0
0 CommissionNo harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
0 Omission No harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 CommissionNo harm
0 Omission No harm
0 Omission No harm
1 Omission Harm
1 Omission Harm
1 CommissionHarm
1 CommissionHarm
0 CommissionNo harm
0 CommissionNo harm
1 CommissionHarm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
1 CommissionHarm
1 Omission Harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
1 Omission Harm
0 CommissionNo harm
1 CommissionHarm
0 Omission No harm
1 CommissionHarm
0 Omission No harm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
1 Omission Harm
0 CommissionNo harm
1 Omission Harm
100
48
100
29
95
32
100
80
55
96
95
95
100
36
64
100
64
90
100
100
41
66
90
100
100
70
92
77
25
91
45
70
100
100
91
100
100
80
22
74
60
100
37
70
38
85
80
0
8.517393
3.931826
6.398595
10.12667
9.903538
0
0
6.908755
11.51294
10.8198
6.216606
8.006701
9.21044
9.615872
8.517393
0
9.903538
0
10.8198
1.791759
9.21044
0
3.258097
8.517393
6.908755
9.21044
6.216606
9.903538
5.70711
8.517393
6.216606
9.21044
7.824446
10.59666
9.21044
0
8.006701
3.931826
15.89597
8.517393
7.601402
8.517393
7.601402
8.517393
7.601402
7.601402
0
5000
50
600
25000
20000
0
0
1000
100000
50000
500
3000
10000
15000
5000
0
20000
0
50000
5
10000
0
25
5000
1000
10000
500
20000
300
5000
500
10000
2500
40000
10000
0
3000
50
8008135
5000
2000
5000
2000
5000
2000
2000
7
6
7
7
7
5
7
7
7
7
5
6
7
6
7
7
6
7
7
7
4
5
6
7
7
6
7
6
6
7
5
6
7
7
6
7
7
6
6
7
6
7
5
7
6
7
5
2
6
3
5
4
5
1
6
2
3
2
6
5
4
3
3
4
4
7
4
4
3
5
6
1
3
5
2
5
5
3
5
7
7
3
4
7
4
5
5
1
1
6
5
6
2
3
1
0
1
1
1
0
0
0
1
0
1
1
0
1
0
0
0
1
1
1
0
0
0
1
0
1
1
0
0
1
1
0
0
0
0
1
1
1
1
0
0
0
0
1
1
1
0
1 CommissionHarm
0 Omission No harm
0 CommissionNo harm
0 CommissionNo harm
1 CommissionHarm
0 Omission No harm
1 Omission Harm
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
0 CommissionNo harm
0 CommissionNo harm
0 Omission No harm
1 CommissionHarm
0 Omission No harm
1 Omission Harm
1 Omission Harm
1 CommissionHarm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
1 Omission Harm
0 Omission No harm
0 CommissionNo harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
1 CommissionHarm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
1 Omission Harm
0 Omission No harm
1 CommissionHarm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
100
6
100
69
39
65
70
55
41
60
85
84
49
80
100
81
47
70
79
74
65
99
76
74
61
100
92
59
97
100
81
60
57
55
90
100
100
76
72
50
30
70
64
76
90
66
80
9.21044
1.791759
0
9.21044
6.216606
7.601402
10.12667
8.517393
10.8198
0
6.908755
0
4.615121
2.397895
3.931826
10.12667
0
3.044522
0
9.21044
8.922792
9.21044
5.303305
10.12667
4.615121
1.791759
11.22526
9.21044
3.044522
3.258097
6.908755
8.699681
0
5.70711
11.51294
9.21044
8.517393
8.853808
7.601402
0
5.303305
6.908755
8.987322
6.908755
10.12667
8.517393
10.12667
10000
5
0
10000
500
2000
25000
5000
50000
0
1000
0
100
10
50
25000
0
20
0
10000
7500
10000
200
25000
100
5
75000
10000
20
25
1000
6000
0
300
100000
10000
5000
7000
2000
0
200
1000
8000
1000
25000
5000
25000
7
7
7
6
7
6
7
7
5
6
7
4
6
7
6
6
4
4
6
7
6
7
7
6
6
7
7
7
7
7
7
6
6
6
7
7
5
6
6
6
5
7
6
7
7
6
7
7
5
6
4
7
5
4
5
3
5
5
4
4
5
1
6
3
2
2
6
5
4
3
2
2
7
7
1
2
7
7
3
4
6
4
6
3
4
3
4
2
4
5
5
5
4
4
0
1
0
1
1
1
0
1
0
0
1
0
1
1
0
0
0
1
0
1
1
0
0
0
1
1
1
0
1
0
0
1
1
0
0
1
0
0
1
1
1
1
0
0
1
0
1
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
0 CommissionNo harm
1 CommissionHarm
1 CommissionHarm
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
1 CommissionHarm
0 Omission No harm
1 Omission Harm
0 Omission No harm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
1 CommissionHarm
1 CommissionHarm
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
1 CommissionHarm
1 Omission Harm
1 CommissionHarm
79
57
51
86
100
34
90
20
0
89
76
83
71
81
81
63
75
0
80
36
93
69
50
75
75
62
95
80
85
100
85
88
100
100
76
65
85
50
71
76
80
60
77
0
100
81
100
2.397895
4.615121
8.517393
3.258097
7.313887
10.38903
6.216606
7.601402
0
9.615872
6.216606
1.791759
8.517393
8.517393
6.216606
9.21044
6.216606
6.216606
6.216606
7.170888
8.517393
0
0
0
10.8198
5.913503
8.517393
6.216606
6.908755
0
6.908755
8.517393
8.006701
0.693147
7.601402
9.21044
3.931826
0
8.517393
10.30899
9.21044
6.621406
8.987322
0
10.59666
10.8198
0
10
100
5000
25
1500
32500
500
2000
0
15000
500
5
5000
5000
500
10000
500
500
500
1300
5000
0
0
0
50000
369
5000
500
1000
0
1000
5000
3000
1
2000
10000
50
0
5000
30000
10000
750
8000
0
40000
50000
0
7
5
5
6
7
4
7
5
7
6
5
3
7
6
7
6
6
5
6
3
6
6
6
7
7
6
7
7
7
4
6
5
7
7
7
7
7
6
6
6
6
5
6
7
6
7
7
3
5
5
4
7
4
4
3
6
3
3
2
4
3
4
3
4
5
6
2
4
5
3
3
2
6
5
7
7
2
4
3
2
1
3
1
5
2
4
6
4
3
6
2
3
4
4
0
0
0
1
1
0
1
0
1
1
0
0
1
0
1
1
0
0
1
0
0
1
1
0
0
1
1
1
1
0
0
1
0
0
1
0
1
1
0
0
0
1
1
0
0
0
0
1 Omission Harm
0 Omission No harm
0 Omission No harm
0 CommissionNo harm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
0 CommissionNo harm
1 CommissionHarm
1 Omission Harm
0 Omission No harm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
1 CommissionHarm
1 CommissionHarm
0 CommissionNo harm
1 CommissionHarm
1 Omission Harm
0 Omission No harm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
0 Omission No harm
0 CommissionNo harm
1 CommissionHarm
1 Omission Harm
0 Omission No harm
1 Omission Harm
0 Omission No harm
70
49
73
81
100
100
84
80
96
91
72
100
91
93
100
43
90
0
74
89
72
75
89
50
67
88
99
0
74
100
49
28
51
80
65
77
100
100
95
65
31
39
76
99
80
0
81
8.517393
4.615121
3.258097
9.21044
3.931826
8.2943
9.615872
0
10.59666
0
9.21044
0
13.81551
11.51294
13.52783
6.908755
9.21044
0
8.517393
10.8198
6.908755
9.21044
7.601402
0
8.2943
13.81551
9.21044
11.51294
10.12667
8.517393
0
7.09091
7.601402
2.397895
7.824446
9.21044
9.21044
6.908755
0
8.517393
0
0
11.51294
6.216606
9.21044
0
0
5000
100
25
10000
50
4000
15000
0
40000
0
10000
0
1000000
100000
750000
1000
10000
0
5000
50000
1000
10000
2000
0
4000
1000000
10000
100000
25000
5000
0
1200
2000
10
2500
10000
10000
1000
0
5000
0
0
100000
500
10000
0
0
2
5
7
7
7
5
6
7
7
6
5
7
4
7
7
4
7
6
5
6
7
6
7
4
6
6
6
7
6
7
7
7
6
6
6
6
7
7
7
6
6
6
7
7
7
6
7
5
4
6
5
7
2
5
2
4
7
4
4
2
5
6
3
4
3
6
5
2
5
5
4
6
7
6
4
6
4
4
5
5
4
5
4
7
6
2
3
5
3
1
1
7
4
4
1
1
1
0
1
1
1
0
1
0
0
0
1
1
0
0
0
1
1
0
1
0
0
0
1
1
0
1
0
1
0
0
1
1
1
0
0
1
0
1
0
1
0
0
1
1
1
1 CommissionHarm
0 CommissionNo harm
0 CommissionNo harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
0 CommissionNo harm
0 Omission No harm
1 CommissionHarm
1 Omission Harm
1 Omission Harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
0 Omission No harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
1 CommissionHarm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
1 Omission Harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
0 CommissionNo harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
1 CommissionHarm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
0 CommissionNo harm
0 Omission No harm
1 Omission Harm
0 CommissionNo harm
1 CommissionHarm
1 CommissionHarm
100
95
100
25
100
100
100
81
90
100
76
84
53
80
83
54
100
100
100
44
80
40
49
69
55
63
80
88
100
72
100
55
81
100
91
0
30
54
44
100
45
27
25
19
61
49
100
10.30899
8.517393
8.517393
6.908755
2.397895
9.21044
0
0
9.21044
10.8198
8.922792
8.517393
6.216606
9.21044
1.791759
8.517393
11.51294
8.517393
7.601402
3.258097
6.908755
0
8.517393
0
9.615872
10.8198
10.12667
10.12667
11.22526
6.216606
9.615872
0
0
10.8198
9.903538
0
0
7.601402
0
2.397895
9.903538
5.303305
8.006701
0
6.216606
6.685861
9.21044
30000
5000
5000
1000
10
10000
0
0
10000
50000
7500
5000
500
10000
5
5000
100000
5000
2000
25
1000
0
5000
0
15000
50000
25000
25000
75000
500
15000
0
0
50000
20000
0
0
2000
0
10
20000
200
3000
0
500
800
10000
7
7
5
6
7
7
7
3
6
6
5
7
4
7
7
6
7
7
6
4
7
5
4
5
6
6
6
5
7
6
7
5
6
7
7
1
6
6
4
7
6
5
5
5
5
7
6
7
5
5
1
7
7
7
2
7
3
3
6
4
5
4
2
5
7
4
5
3
2
4
5
5
5
3
5
1
5
7
5
4
7
2
1
3
6
4
5
3
6
4
3
6
7
3
0
1
0
1
0
1
0
1
0
0
0
1
1
0
1
1
0
0
1
0
1
1
0
0
0
1
1
0
1
0
1
0
0 Omission No harm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
1 Omission Harm
0 CommissionNo harm
0 Omission No harm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
0 Omission No harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
1 CommissionHarm
0 CommissionNo harm
1 Omission Harm
0 Omission No harm
1 CommissionHarm
1 Omission Harm
0 CommissionNo harm
1 CommissionHarm
1 Omission Harm
0 Omission No harm
0 Omission No harm
0 CommissionNo harm
1 CommissionHarm
1 Omission Harm
1 CommissionHarm
0 Omission No harm
0 CommissionNo harm
1 Omission Harm
22
100
10
83
76
100
17
80
95
64
70
100
60
100
100
100
60
100
81
100
59
92
37
30
33
71
86
90
51
55
34
100
6.908755
6.908755
7.313887
5.303305
9.615872
3.931826
0
0
7.313887
5.303305
0
9.21044
6.216606
5.525453
8.517393
6.908755
0
8.517393
9.903538
9.615872
6.908755
9.21044
0
8.006701
0
0
9.903538
0
9.903538
0
2.397895
9.21044
1000
1000
1500
200
15000
50
0
0
1500
200
0
10000
500
250
5000
1000
0
5000
20000
15000
1000
10000
0
3000
0
0
20000
0
20000
0
10
10000
7
7
7
7
7
6
6
7
4
6
3
7
6
7
7
7
7
6
7
7
5
7
4
6
6
6
6
7
4
6
6
7
7
6
7
6
6
4
2
7
4
1
2
2
4
4
7
3
5
3
5
7
4
6
3
4
2
6
5
2
4
5
2
7
Regret
Age
1
3
4
6
5
5
1
2
3
2
2
2
1
4
4
3
1
5
3
1
5
4
3
2
2
3
4
4
4
5
2
4
2
4
2
6
1
2
2
5
2
5
4
4
4
3
Gender
23
25
26
33
31
71
25
34
27
27
25
18
29
27
29
37
25
22
54
21
29
72
53
25
21
25
51
34
41
38
40
32
27
39
23
46
40
54
37
54
22
18
29
54
34
28
1
2
1
2
2
1
1
1
1
2
2
2
2
1
2
1
1
1
1
1
1
2
2
1
2
1
2
2
2
2
2
2
1
1
1
2
2
1
2
1
2
1
1
2
1
2
4
6
4
4
4
4
2
1
2
3
1
4
4
2
4
2
4
4
1
4
4
3
4
1
1
3
2
4
6
2
6
3
1
4
4
1
1
2
6
3
1
1
6
1
2
4
5
44
24
44
26
36
34
55
30
28
34
33
40
34
31
56
26
21
23
49
47
24
39
26
35
32
56
44
60
25
29
26
48
31
31
32
44
67
28
28
26
39
29
26
61
21
53
58
2
2
2
1
1
2
2
1
1
2
1
2
2
2
1
2
1
1
2
2
2
1
1
2
2
2
2
2
1
1
1
1
2
1
2
2
2
1
1
1
2
2
1
1
1
2
1
1
1
4
4
2
2
2
2
3
3
4
4
4
3
4
2
4
3
5
1
2
4
4
2
5
7
1
3
4
1
2
3
2
3
1
1
1
4
2
2
5
2
2
2
1
3
3
48
27
62
44
46
42
36
25
40
48
35
28
31
44
26
24
28
28
61
34
40
34
30
51
62
56
42
33
48
28
51
41
26
19
31
26
35
39
34
51
30
30
24
19
36
50
53
2
2
2
2
1
1
2
1
1
2
1
2
2
2
2
1
2
1
2
2
1
1
2
2
1
2
1
1
2
1
2
2
1
1
2
2
2
2
2
1
1
2
1
2
2
1
1
4
3
5
4
1
5
4
2
1
4
2
5
5
4
4
4
4
3
4
4
2
3
2
2
2
1
1
1
2
1
3
2
4
1
2
1
2
4
4
2
4
4
2
1
2
2
2
35
29
56
36
37
28
27
33
22
34
21
30
31
30
35
26
55
29
33
25
46
33
22
33
44
45
28
29
36
54
33
23
49
34
28
38
30
25
58
32
24
24
51
29
38
40
54
2
2
2
2
1
1
2
2
1
1
2
1
2
1
2
1
1
1
1
1
1
1
2
2
2
1
1
2
1
1
2
1
2
2
2
1
1
2
2
1
1
1
1
1
2
2
2
4
5
5
1
1
2
5
2
1
1
4
1
6
1
4
4
4
3
4
1
5
4
3
4
2
1
2
4
6
3
2
4
2
2
4
4
6
4
4
2
4
4
6
6
4
4
4
29
30
41
35
46
29
36
27
51
70
30
62
30
32
36
25
44
31
31
59
42
37
34
27
29
29
74
36
40
30
64
32
73
24
39
35
23
42
44
36
38
25
37
41
25
43
53
2
1
2
1
2
2
2
1
2
2
2
1
2
2
2
1
2
2
2
2
2
1
1
1
1
1
1
1
1
1
2
1
1
2
1
2
1
2
2
2
2
1
1
1
1
2
2
1
2
4
4
1
1
1
4
2
1
5
4
4
4
3
3
1
1
2
5
4
5
6
3
4
3
4
1
1
4
1
4
4
1
4
7
2
4
4
2
4
4
4
3
4
2
1
52
64
29
28
24
35
70
61
34
38
51
31
27
36
29
24
33
35
34
30
34
48
40
30
39
26
32
20
32
41
51
31
38
55
21
29
30
29
30
24
25
27
27
76
32
41
43
2
1
1
1
2
1
2
1
1
2
2
1
1
1
2
2
1
2
2
1
1
2
1
2
2
1
1
1
1
1
2
2
1
2
1
1
2
1
2
2
1
2
1
1
1
2
2
1
1
1
1
2
1
3
5
5
1
2
1
3
1
1
1
4
1
2
1
3
2
3
2
4
1
1
4
4
4
2
1
28
30
30
43
32
50
35
28
23
31
44
45
44
45
35
48
28
33
28
31
59
30
36
53
46
33
25
22
21
43
18
28
2
2
2
2
2
2
1
1
1
1
2
1
2
1
1
1
2
1
2
1
1
2
1
1
1
2
2
1
1
2
1
1