Replication and Extension of Alicke (1985) Better-Than-Average Effect for Desirable and ‎Controllable Traits

Abstract

People tend to regard themselves as better than average. We conducted a replication and ‎extension of Alicke's (1985) classic study on trait dimensions in evaluations of self versus ‎others with U.S. American MTurk workers in two waves (total N = 1573; 149 total traits). ‎We successfully replicated the trait desirability effect, such that participants rated more ‎desirable traits as being more descriptive of themselves than of others (original: ηp2 = .78, ‎‎95% CI [.73, .81]; replication: sr2 = .54, 95% CI [.43, .65]). The effect of desirability was ‎stronger for more controllable traits (effect of desirability X controllability interaction on ‎self-other ratings difference, original: ηp2 = .21, 95% CI [.12, .28]; replication: sr2 = .07, ‎‎95% CI [.02, .12]). In an extension, we found that desirable traits were rated as more ‎common for others, but not for the self. Thirty-five years later, the better-than-average ‎effect appears to remain robust.‎

Full text

Replication and Extension of Alicke (1985) Better-Than-Average Effect
for Desirable and Controllable Traits
*Ignazio Ziano
Department of Marketing, Grenoble Ecole de Management,
Univ Grenoble Alpes ComUE
ignazio.ziano@grenoble-em.com
* Pui Yan (Cora) Mok
Department of Psychology, University of Hong Kong, Hong Kong SAR
coramok@gmail.com
^*Gilad Feldman
Department of Psychology, University of Hong Kong, Hong Kong SAR
gfeldman@hku.hk / giladfel@gmail.com
*Contributed equally, joint first author
^Corresponding author
Word: abstract – [149], manuscript - [4870] (excluding tables/figures)

Corresponding author
Gilad Feldman, Department of Psychology, University of Hong Kong, Hong Kong SAR;
gfeldman@hku.hk
Author bios:
Gilad Feldman is an assistant professor with the University of Hong Kong psychology
department. His research focuses on judgment and decision-making.
Ignazio Ziano is an assistant professor at the Department of Marketing, Grenoble Ecole de
Management, Univ Grenoble Alpes ComUE. His research focuses on consumer behaviour and
judgment and decision-making.
Mok Pui Yan (Cora) was a guided thesis master’s student working under the supervision of
Gilad Feldman at the University of Hong Kong in the academic year 2018-2019.
Declaration of Conflict of Interest:
The author(s) declared no potential conflicts of interests with respect to the authorship and/or
publication of this article.
Financial disclosure/funding:
The research was supported by the European Association for Social Psychology seedcorn grant
awarded to the corresponding author.
Authorship declaration:
Cora conducted this project for her master's thesis. She initiated and designed the studies, wrote
the pre-registration, ran the initial analyses, and wrote the initial draft as her master’s thesis.
Gilad was the advisor, supervised each step in the project, conducted the pre-registrations, and
ran data collections. Ignazio followed up on initial work by Cora and Gilad to verify analyses
and conclusions, and completed the manuscript submission draft. Ignazio and Gilad jointly
finalized the manuscript for submission and handled revisions.

In the table below, we employ CRediT (Contributor Roles Taxonomy) to identify the
contribution and roles played by the contributors in the current replication effort. Please refer to
the url (https://www.casrai.org/credit.html ) on details and definitions of each of the roles listed
below.
Role
Ignazio
Ziano
Pui Yan (Cora)
Mok
Gilad
Feldman
Conceptualization
X
X
Pre-registration
X
Data curation
X
Formal analysis
X
X
Funding acquisition
X
Investigation
X
X
X
Methodology
X
X
Pre-registration peer
review / verification
X
X
Data analysis peer
review / verification
X
X
Project administration
X
Resources
X
Supervision
X
Validation
X
X
Visualization
X
X
Writing-original draft
X
X
Writing-review and
editing
X
X

Replication and extension of Alicke (1985) 1
Abstract
People tend to regard themselves as better than average. We conducted a replication and
extension of Alicke's (1985) classic study on trait dimensions in evaluations of self versus
others with U.S. American MTurk workers in two waves (total N = 1573; 149 total traits).
We successfully replicated the trait desirability effect, such that participants rated more
desirable traits as being more descriptive of themselves than of others (original: ηp2 = .78,
95% CI [.73, .81]; replication: sr2 = .54, 95% CI [.43, .65]). The effect of desirability was
stronger for more controllable traits (effect of desirability X controllability interaction on
self-other ratings difference, original: ηp2 = .21, 95% CI [.12, .28]; replication: sr2 = .07, 95%
CI [.02, .12]). In an extension, we found that desirable traits were rated as more common for
others, but not for the self. Thirty-five years later, the better-than-average effect appears to
remain robust.
Keywords: better-than-average effect, self-evaluation, comparative judgment, replication

Replication and extension of Alicke (1985) 2
Replication and Extension of Alicke (1985) Better-Than-Average Effect
for Desirable and Controllable Traits
People seem to regard themselves as better than average in many domains. When
asked to compare themselves with the average other, people tend to rate themselves
possessing more positive traits, being better drivers, and engaging in more desirable
behaviors such as contributing to charity (Brown, 2012; Epley & Dunning, 2000; Svenson,
1981). This better-than-average effect - the tendency to evaluate oneself more favorably than
the average other person - has received wide attention in the social psychology literature
(Alicke & Govorun, 2005; Krueger & Mueller, 2002).
The better-than-average effect has implications for human decision-making and
judgment. People often make decisions based on how they view themselves in comparison to
the average other person. Such self-evaluation may concern their skills, personal attributes or
even physical conditions, thus influencing many domains of life, including education, health,
business, and sports (Dunning, Heath, & Suls, 2004; Guenther, Taylor, & Alicke, 2015;
Malmendier, & Tate, 2005; Stanley et al., 2017; Taylor & Brown, 1988; Ziano & Villanova,
2019). If their evaluation is indeed inaccurate, it is necessary to understand the process
behind the phenomenon.
There are two types of explanations for the better-than-average effect. The
motivational explanation argues that the phenomenon is a type of self-enhancement for
people to protect and maintain their self-worth (Alicke, Zell, & Guenther, 2013; Sedikides,
Gaertner, & Toguchi, 2003). On the other hand, the non-motivational explanation suggests
the better-than-average effect arises from biases in information processing. It may be easier
for people to evaluate a single object than an abstract entity like the average other, which can
lead to inaccurate comparative judgment in the better-than-average paradigm (Chambers &
Windschitl, 2004; Krizan & Suls, 2008), and the vagueness of the scale may also play a part

Replication and extension of Alicke (1985) 3
(Logg, Haran, & Moore, 2018), such that better-than-average effects are stronger when the
scale is vague and leaves some space for arbitrary interpretation, compared to when the scale
is more concrete. While both interpretations may be relevant, researchers have yet to identify
a more parsimonious explanation that reconciles them.
Choice of study for replication
We aimed to conduct a direct replication of Alicke (1985), one of the classic studies
on the better-than-average effect. We selected Alicke (1985) for several reasons. One is its
academic impact. Published more than three decades ago, the study is one of the earliest
attempts to demonstrate the better-than-average effect. At the time of writing, it had more
than 1100 citations according to Google Scholar, including those by prominent review papers
and textbooks (Brown, 2014; Dunning, Heath, & Suls, 2004; Mischel, Shoda, & Ayduk,
2007; Taylor & Brown, 1988). Second, to the best of our knowledge there are no direct
replications, but only conceptual replications of the study . Building on the findings from
Alicke (1985), some studies have found support for the better-than-average effect such that
people tended to regard more positive traits as more descriptive of themselves than of the
average other (Brown, 2012; Kanten & Teigen, 2008; Pedregon et al., 2012). However,
conceptual replications alone cannot verify the robustness of the original findings (Simmons
et al., 2011), as differences in procedure and stimuli could cause discrepant results. Direct
replications can fill this gap. By operationalizing the variables in the same way as the
original, they may help retest these findings, and examine whether they are solid foundations
for building and strengthening theories (Zwaan, Etz, Lucas, & Donnellan, 2018).
Alicke (1985) conducted two data collections. In the first, he asked undergraduates in
to rate various traits in terms of their desirability and controllability, and these were used to
form categories of desirability and controllability for an experiment. Participants in a second

Replication and extension of Alicke (1985) 4
sample were asked to rate how well these traits characterize them and the average college
student.
Alicke's findings revealed that participants rated more desirable traits as more
characteristic of themselves than the average student. Further, when the traits were more
desirable, participants believed that traits of higher controllability were even more
characteristic of themselves compared to others, generating a desirability by controllability
interaction on self-minus-other ratings.
Replication and extension
We planned to revisit the original findings with two replication hypotheses, and
extend the original article with one extension hypothesis. We summarized the hypotheses in
the present study in Table 1.
Table 1
Summary of replication and extension hypotheses
Replication Hypotheses
1
The difference between evaluation of self and others is higher as traits increase in
desirability.
2
Among high desirable traits, self-ratings are higher than other-ratings for high
controllable traits than for low-controllable traits, whereas among low-desirable
traits, self-ratings are higher than other-ratings for low-controllable traits than for
high-controllable traits.
Extension Hypothesis
3
For ratings of others, trait desirability is positively associated with trait
commonness. For ratings of self, trait desirability is negatively associated with
trait commonness.
The extension investigated the role of desirability and rating perspective on
commonness (i.e., how widespread in the population the trait is perceived to be), with the

Replication and extension of Alicke (1985) 5
goal to address the methodological concerns around the better-than-average effect. Some
researchers have argued that the effect may result from the ambiguous criteria involved in
comparative judgments between the self and the average other person (Dunning, Meyerowitz,
& Holzberg, 1989). Such ambiguity might leave the criteria open to participants’
interpretation and thus confound results. A similar argument addressing Alicke’s (1985)
study was that it may have confounded trait desirability with commonness (Moore, 2007). It
suggested that people might report perceiving traits like friendliness as more self-descriptive
as these traits were likely displayed more often than those like rudeness or dishonesty. In this
regard, trait desirability may be confounded with trait commonness. On the other hand,
research has suggested that people with higher self-esteem perceive their desirable traits as
less common (Ditto & Griffin, 1993). If people are motivated to enhance their self-view as
the better-than-average effect would postulate, there is a reason to believe that the
relationship between desirability and commonness may be dependent on the rating
perspective, such that they would find desirable traits with higher self-ratings less common,
and those with higher other-ratings more common.
Adjustments to the original study
We made four adjustments to the original procedure.
First, we changed the design of the second data collection. After completing the first
data collection, we conducted an initial analysis of the results, with the goal to categorize
traits into four levels of desirability and two levels of controllability. However, results from
the first sample prompted a departure from this plan. Only slight decimal differences were
observed in the ratings of desirability and controllability, which would pose a challenge to
categorizing the variables into meaningful levels. Additionally, categorizing continuous
predictors may weaken the ability to detect actual relationships (Irwin & McClelland, 2003;
MacCallum, Zhang, Preacher, & Rucker, 2002). To examine the relationships between the

Replication and extension of Alicke (1985) 6
variables more meaningfully, we decided to change the second data collection: instead of
assigning participants to specific levels of desirability and controllability, we randomly
assigned participants to one of the three conditions: ratings from the self-perspective, from
the average American perspective, or in terms of trait commonness.
Second, whereas in the original paper the same participants rated themselves and the
average students (a within-subjects design for the self-other ratings), we had one group of
participants rating themselves and another group rating the average American (a between-
subjects design).
Third, we conducted the final analysis on an item level. From all participant ratings,
we calculated the mean for each trait on each dimension. To validate the change in our
planned analysis, we tested the item-level analysis, using the data obtained in the first data
collection and on a randomly simulated dataset on the planned for the second data collection.
Fourth, we addressed the rating perspective by examining the effects of desirability
and controllability on self-minus-other ratings, instead of treating it as a predictor. The
decision to examine only two-way interactions aimed to improve the clarity of interpretation
in statistical analyses. The change helped address the issue of drawing inferences about the
relative importance of multiple two-way interactions in a complex three-way analysis of
variance, given their differential levels of residual variance (McClelland & Judd, 1993).

Replication and extension of Alicke (1985) 7
Pre-registration and Open Science
For the two data collections, we first pre-registered the experiment on the Open
Science Framework (OSF) and data collection was launched soon after. Pre-registrations,
disclosures, power analyses, and all materials are available in the supplementary. These,
together with datasets and R/RMarkdown code, were made available on the OSF at
https://osf.io/2y6wj/?view_only=24557c4eea50418cbf07a858c09c5d1c . All measures,
manipulations, and exclusions for this investigation are reported, and data collection was
completed before analyses.
Pre-registrations are all available on the OSF
1
: First data collection pre-registration -
https://osf.io/fyzwd/?view_only=285a1615216c4ab68185c1b14ede11ba ; Updated second
data collection pre-registration following first data collection insights:
https://osf.io/9esva/?view_only=024fbb533e594903b586a93187f9e7cb .
Method
Power Analysis
We pre-registered a power analysis of the results described in Alicke (1985) and
included the analysis in the supplementary materials (α = .05, one-tailed, power = .95;
G*Power 3.1.9.3). Based on the original reported test statistics of the Desirability ×
Controllability interaction, F(3, 261) = 22.72, we calculated an effect size estimate of ηp2 =
0.21, 95% CI [.12, .28]. With this estimate, a minimum of 71 participants was required to
1
We note that prior to the final pre-registration of Wave 1 we had two prior pre-registrations which we found
had to be amended due to issues identified in the comprehension checks and a Qualtrics bug that affected
randomization. Amendments were made prior to the full first data collection. Links to prior registrations:
https://osf.io/a5mx7/?view_only=7b211abff2ee4ba380c1ef4b093fcbdf and
https://osf.io/pvr6t/?view_only=debd871aceca4db5bd2b035e0927d2b2 . The final pre-registration for the first
data collection was completed before data collection. In addition, we already pre-registered the second data
collection in the pre-registration of the first data collection, yet following our analysis of the first data collection
we made changes to the pre-registration of the section data collection. These changes are explained in the
"adjustments to the original study" section above. For the most complete pre-registration plan conducted prior to
data collections please refer to the latest pre-registrations.

Replication and extension of Alicke (1985) 8
achieve 95% power with an α-level of .05 for each condition. Having also taken into account
the switch to a between-subjects design, which typically have lower statistical power than
within-subjects designs, we aimed for at least 640 participants for the initial ratings, and at
least 894 participants for the second sample.
Participants and Procedure
Both the first and second samples were recruited online via Amazon Mechanical Turk
(MTurk) using TurkPrime.com (Litman, Robinson, & Abberbock, 2017) in return for $0.50
(estimated completion time ~4 minutes, to meet minimum federal wage of $7.25). The first
sample comprised of 670 participants, who rated the degree of desirability (n = 329) or
controllability (n = 341) of traits to the average American. The second sample comprised of
903 participants, who rated the degree to which these traits characterized themselves (n =
300) or the average American (n = 306), or the degree of commonness of these traits to the
average American (n = 297). Six participants, four in the first sample and two in the second
sample, were excluded from the analyses since they were detected to be based outside the
United States and therefore were not allowed to proceed and answer the questionnaire. A
comparison between the study characteristics between the original study and the replication is
summarized in Table S11 – S12 in the supplementary materials.
We did not have access to an American undergraduate student population (recruited
for the original study) for this replication. We used MTurk samples because of the
convenience MTurk provides in reaching a large enough sample size in a short time. MTurk
samples have been shown to produce very similar results to U.S. representative samples in
experimental political psychology (Coppock, 2017; Coppock, Leeper, & Mullinix, 2018;
Mullinix, Leeper, Druckman, & Freese, 2015). Coming to social psychology results, there are
several examples of replication of studies originally conducted with U.S. American
undergraduate students which were successfully replicated with MTurk. For instance,

Replication and extension of Alicke (1985) 9
overestimation of others’ willingness-to-pay (Frederick, 2012) was successfully replicated on
MTurk (Jung, Moon, & Nelson, 2019, study 3). An ongoing mass replication effort
successfully replicated a large number of judgment and decision making studies using
Amazon Mechanical Turk, with results consistent with student samples and other online
recruitment platforms such as Prolific (Collaborative Open-science Research, 2020;
Chandrashekar et al., 2019; Chen et al., 2020; Ziano et al., 2020). Overall, this supports
MTurk as a viable sample for a replication of Alicke (1985).
The surveys for both the first and second samples were conducted online using
Qualtrics. Participants were randomly assigned to one of the conditions. They then received
instructions about the rating criteria of their assigned condition and answered comprehension
questions accordingly. After answering these questions, they were asked to evaluate 40 traits,
randomized out of the 149 traits derived from Alicke's (1985) study.
Design and Analyses
The present study is a between-subjects design with three independent variables (trait
desirability, commonness, and controllability) and two dependent variables (self- and other-
ratings). Analyses were conducted on an item level by averaging all participant ratings on
each dimension for each trait.
To account for the rating perspective in the replication hypotheses, we calculated self-
minus-other ratings by subtracting other-ratings from self-ratings. A positive value indicates
that participants perceived the specific trait as more characteristic of themselves than of the
average other whereas a negative value indicates the trait is regarded as less characteristic of
themselves than of the average other. For both data collections, details about attention checks
and exclusion criteria are available in the supplementary materials.
Materials
First data collection

Replication and extension of Alicke (1985) 10
Before being able to proceed with the survey, participants were asked three
comprehension checks (described in details in the Supplementary materials).
Desirability.
A desirable characteristic was defined as something the average American perceives
as good to have and an undesirable characteristic as something the average American
perceives as bad to have. This definition was identical to that in Alicke (1985), except that
the original reference point “average college student” was replaced by “average American” in
order to cater to the participant population in the present study. Participants rated to what
extent a trait was desirable (1 = very undesirable; 7 = very desirable).
Controllability.
A controllable characteristic was defined as something that an average American can
create or eliminate with a sufficient amount of effort, whereas an uncontrollable characteristic
was something that an average American’s effort would not suffice to create or eliminate.
This definition was identical to that in Alicke (1985), except that the original reference point
“average college student” was replaced by “average American” in order to cater to the
participant population in the present study. Participants rated to what extent a trait was
controllable on a scale from 1 to 7 (1 = very uncontrollable; 7 = very controllable).
Second data collection
Before being able to proceed with the survey, participants were asked three
comprehension checks (described in detail in the Supplementary materials).
Commonness.
A common characteristic was defined as one that an average American frequently
displays, whereas an uncommon characteristic was defined as something that an average
American rarely displays. This definition of commonness was taken from Moore’s (2007)
review paper, which argues that this dimension was a potential confound with desirability in

Replication and extension of Alicke (1985) 11
the original study. Participants rated to what extent a trait was common (1 = very uncommon;
7 = very common).
Traits.
A total of 149 traits were used in the present study. These traits were originally
derived from Anderson (1968) and are identical to the final list reported in the appendix of
Alicke (1985). Although the study reported using 154 traits, a detailed examination of the list
revealed a total number of only 149. In the present study, participants in each condition rated
40 of these traits in randomized order. These traits are summarized in the supplementary.
For self-ratings, participants rated to what extent a trait characterized themselves (1 =
not at all characteristic of me; 7 = very characteristic of me). For others' ratings participants
rated to what extent a trait characterized the average American (1 = not at all characteristic
of the average American; 7 = very characteristic of the average American).
Classification of Replication
The replication was identical to the original in terms of the operationalization and
stimuli used for both the independent variable (IV) and the dependent variable (DV). It
differed from the original in the procedural details, physical settings, and contextual
variables. According to LeBel et al.’s (2018) taxonomy, the present study meets the criteria
for a close replication (see Table 2).

Replication and extension of Alicke (1985) 12
Table 2
Classification of replication based on LeBel et al.’s (2018) taxonomy
Design Facet
Replication
Independent variable operationalization
Same
Dependent variable operationalization
Same
Independent variable stimuli
Same
Dependent variable stimuli
Same
Procedural details
Different
Physical settings
Different
Contextual variables
Different
Results
We summarized means, standard deviations, and correlations in Table 3, and the
means and standard deviations of each dimension for all traits in Table 4. To investigate the
relationships between the types of ratings, we performed correlation analyses, correlation
comparisons, and multiple linear regression analyses on the item level. In regression
analyses, all variables were centered for calculating the interaction term to avoid the problem
of multi-collinearity (Aiken, West, & Reno, 1991). The significance level was defined by p
< .05, one-tailed test for replication hypotheses, two-tailed test for the extension hypothesis.
To determine the relative magnitude of each predictor, we used squared semi-partial
correlation coefficient, to address unique variance explained by the specific predictor when
holding other predictors constant. In line with the original study, we calculated an additional
dependent variable (self-other ratings) by subtracting other-ratings from self-ratings.

Replication and extension of Alicke (1985) 13
Table 3
Means, standard deviations, and correlations with confidence intervals
Variable
M
SD
Desirability
Controllability
Commonness
Self-ratings
Other-ratings
Desirability
3.73
1.78
Controllability
4.86
0.81
.03
[-.13, .19]
Commonness
4.07
0.54
.64**
.22**
[.54, .73]
[.06, .37]
Self-ratings
3.73
1.28
.92**
.04
.61**
[.89, .94]
[-.12, .20]
[.50, .70]
Other-ratings
3.97
0.58
.61**
.14
.92**
.55**
[.50, .70]
[-.02, .30]
[.89, .94]
[.42, .65]
Self-minus-other
ratings
-0.24
1.08
.77**
-.03
.23**
.89**
.11
[.69, .82]
[-.19, .13]
[.07, .37]
[.86, .92]
[-.05, .27]
Note. M and SD are used to represent mean and standard deviation, respectively. Values in square brackets indicate the 95% confidence interval
for each correlation. * indicates p < .05. ** indicates p < .01. Analyses were conducted on an item level. Ratings of desirability and
controllability were collected in the first sample, and those of commonness, self-ratings, and other-ratings were collected in the second sample.
Self-minus-other represents self-ratings deducted by other-ratings.

Replication and extension of Alicke (1985) 14
Table 4
Mean ratings and standard deviations of traits in terms of desirability, controllability and
commonness, self-ratings, and other-ratings
Desirability
(N = 149)
Controllability
(N = 149)
Commonness
(N = 149)
Self
(N = 149)
Other
(N = 149)
Traits
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
Intelligent
6.41 (1.02)
4.01 (1.81)
4.57 (1.24)
5.39 (1.35)
4.55 (1.16)
Reliable
6.40 (1.01)
5.93 (1.13)
4.76 (1.31)
5.90 (1.08)
4.57 (1.09)
Loyal
6.33 (1.00)
5.69 (1.42)
4.84 (1.20)
5.93 (1.34)
4.73 (1.10)
Attractive
6.33 (0.97)
3.35 (1.62)
4.32 (1.19)
4.31 (1.39)
4.35 (1.08)
Responsible
6.31 (1.16)
5.82 (1.32)
4.65 (1.20)
5.74 (1.26)
4.67 (1.31)
Resourceful
6.27 (1.01)
5.13 (1.49)
4.8 (1.14)
5.48 (1.12)
4.81 (1.43)
Kind
6.26 (1.08)
5.72 (1.40)
4.8 (1.20)
5.56 (1.19)
4.53 (1.14)
Sincere
6.2 (1.01)
5.33 (1.62)
4.59 (1.14)
5.76 (1.09)
4.51 (1.17)
Friendly
6.16 (1.01)
5.88 (1.19)
4.75 (1.16)
5.51 (1.18)
4.89 (1.35)
Dependable
6.15 (1.33)
5.64 (1.32)
4.78 (1.28)
5.85 (1.27)
4.62 (1.11)
Respectful
6.15 (1.03)
5.97 (1.10)
4.37 (1.37)
5.85 (0.98)
4.15 (1.41)
Admirable
6.12 (1.03)
4.8 (1.68)
4.29 (1.26)
4.38 (1.45)
4.44 (1.36)
Wise
6.09 (1.25)
3.73 (1.62)
4.08 (1.25)
4.85 (1.44)
3.89 (1.34)
Good-tempered
6.09 (1.13)
4.89 (1.70)
4.67 (1.27)
5.10 (1.37)
4.52 (1.28)
Interesting
6.09 (0.97)
4.08 (1.67)
4.37 (1.21)
4.78 (1.34)
4.83 (1.3)
Bright
6.07 (1.00)
4.06 (1.81)
4.45 (1.23)
5.60 (1.21)
4.45 (1.23)
Honorable
6.04 (1.13)
5.30 (1.74)
4.38 (1.28)
5.32 (1.39)
4.49 (1.32)
Clear-headed
6.02 (1.23)
4.58 (1.52)
4.23 (1.28)
5.03 (1.54)
4.11 (1.17)
Pleasant
6.01 (1.20)
5.46 (1.41)
5.03 (1.10)
5.49 (1.07)
4.52 (1.10)
Ethical
6.01 (1.16)
5.58 (1.49)
4.51 (1.14)
5.54 (1.40)
4.33 (1.27)
Level-headed
5.99 (1.26)
4.71 (1.59)
4.43 (1.24)
5.14 (1.51)
4.32 (1.29)
Intellectual
5.99 (1.16)
4.06 (1.79)
4.19 (1.33)
5.06 (1.17)
3.92 (1.17)
Considerate
5.98 (1.37)
5.99 (1.06)
4.28 (1.41)
5.75 (1.08)
4.47 (1.30)
Self-disciplined
5.98 (1.26)
5.59 (1.24)
4.14 (1.29)
4.70 (1.61)
3.95 (1.32)
Polite
5.98 (1.16)
6.36 (0.87)
4.56 (1.19)
5.65 (1.15)
4.12 (1.36)
Punctual
5.97 (1.04)
6.27 (1.03)
4.41 (1.31)
5.57 (1.56)
4.24 (1.31)

Replication and extension of Alicke (1985) 15
Table 4 (Continued)
Mean ratings and standard deviations of traits in terms of desirability, controllability and
commonness, self-ratings, and other-ratings
Desirability
(N = 149)
Controllability
(N = 149)
Commonness
(N = 149)
Self
(N = 149)
Other
(N = 149)
Traits
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
Versatile
5.95 (1.13)
4.87 (1.38)
4.71 (1.17)
4.65 (1.55)
4.55 (1.29)
Clean
5.91 (1.11)
6.20 (1.29)
4.85 (0.95)
5.47 (1.47)
4.68 (1.31)
Humorous
5.82 (1.17)
4.19 (1.54)
4.73 (1.18)
5.22 (1.41)
4.62 (1.10)
Original
5.81 (1.20)
4.02 (1.64)
4.12 (1.26)
4.80 (1.42)
3.98 (1.51)
Grateful
5.80 (1.29)
5.86 (1.34)
4.33 (1.34)
5.30 (1.50)
4.17 (1.51)
Trustful
5.77 (1.24)
5.14 (1.57)
4.78 (1.11)
5.21 (1.46)
4.24 (1.2)
Persistent
5.77 (1.17)
5.68 (1.36)
4.77 (1.23)
5.11 (1.52)
4.79 (1.28)
Lucky
5.76 (1.24)
1.85 (1.45)
4.12 (1.17)
3.43 (1.64)
4.19 (1.30)
Mature
5.71 (1.40)
4.93 (1.70)
4.26 (1.06)
5.32 (1.29)
4.00 (1.14)
Perceptive
5.71 (1.22)
4.03 (1.70)
4.31 (1.16)
5.62 (1.36)
4.27 (1.43)
Sharp-witted
5.71 (1.19)
3.77 (1.81)
4.04 (1.40)
4.92 (1.57)
4.40 (1.26)
Creative
5.64 (1.27)
3.81 (1.63)
4.45 (1.21)
4.79 (1.62)
4.31 (1.45)
Cooperative
5.63 (1.38)
6.17 (1.09)
4.83 (1.34)
5.51 (1.04)
4.57 (1.38)
Observant
5.63 (1.23)
5.42 (1.39)
4.52 (1.22)
5.61 (1.43)
4.29 (1.56)
Lively
5.61 (1.17)
4.57 (1.54)
4.97 (1.21)
4.50 (1.56)
4.91 (0.93)
Clever
5.60 (1.41)
3.89 (1.78)
4.13 (1.38)
5.29 (1.29)
4.22 (1.36)
Imaginative
5.60 (1.26)
3.71 (1.70)
4.61 (1.27)
5.17 (1.41)
4.29 (1.28)
Sportsmanlike
5.55 (1.26)
5.59 (1.47)
4.42 (1.13)
4.67 (1.86)
4.48 (1.25)
Neat
5.54 (1.09)
5.98 (1.16)
4.33 (1.14)
4.84 (1.57)
3.93 (1.28)
Normal
5.46 (1.29)
4.24 (1.58)
5.28 (1.23)
5.12 (1.54)
5.00 (1.12)
Witty
5.40 (1.35)
3.72 (1.71)
4.29 (1.20)
4.69 (1.73)
4.20 (1.31)
Well read
5.33 (1.34)
5.80 (1.35)
3.84 (1.45)
5.09 (1.51)
3.76 (1.66)
Fearless
5.26 (1.42)
3.81 (1.73)
3.86 (1.53)
3.35 (1.6)
3.99 (1.58)
Bold
5.25 (1.17)
4.79 (1.49)
4.59 (1.13)
3.72 (1.52)
4.73 (1.17)
Quick
5.08 (1.22)
3.94 (1.70)
4.26 (1.19)
4.49 (1.43)
4.31 (1.32)
Fashionable
5.04 (1.16)
5.72 (1.37)
4.46 (1.08)
3.58 (1.84)
4.01 (1.32)
Progressive
5.00 (1.28)
5.28 (1.41)
4.51 (1.25)
4.83 (1.72)
4.40 (1.22)
Ingenious
4.96 (1.79)
3.60 (1.77)
3.69 (1.51)
3.94 (1.75)
3.90 (1.49)

Replication and extension of Alicke (1985) 16
Table 4 (Continued)
Mean ratings and standard deviations of traits in terms of desirability, controllability and
commonness, self-ratings, and other-ratings
Desirability
(N = 149)
Controllability
(N = 149)
Commonness
(N = 149)
Self
(N = 149)
Other
(N = 149)
Traits
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
Self-satisfied
4.73 (1.57)
4.81 (1.57)
4.62 (1.38)
3.91 (1.66)
4.81 (1.27)
Thrifty
4.70 (1.37)
5.59 (1.48)
3.91 (1.48)
4.96 (1.59)
3.54 (1.36)
Philosophical
4.70 (1.34)
4.48 (1.66)
3.33 (1.45)
4.36 (1.70)
3.43 (1.50)
Prudent
4.67 (1.48)
4.91 (1.50)
4.04 (1.22)
4.38 (1.72)
4.07 (1.42)
Religious
4.62 (1.35)
5.66 (1.56)
4.73 (1.41)
3.21 (2.18)
4.43 (1.31)
Meticulous
4.59 (1.37)
5.26 (1.56)
3.77 (1.06)
4.29 (1.81)
3.70 (1.24)
Obedient
4.52 (1.44)
6.00 (1.16)
4.26 (1.22)
4.45 (1.41)
4.01 (1.44)
Authoritative
4.5 (1.44)
4.91 (1.44)
4.54 (1.21)
3.54 (1.88)
4.22 (1.37)
Changeable
4.36 (1.25)
5.30 (1.62)
4.20 (1.26)
4.04 (1.38)
4.25 (1.38)
Sensitive
4.35 (1.40)
3.92 (1.79)
4.35 (1.22)
4.96 (1.54)
4.42 (1.23)
Conforming
4.11 (1.52)
5.20 (1.41)
4.86 (1.26)
3.94 (1.52)
4.48 (1.37)
Reserved
4.10 (1.38)
4.69 (1.76)
3.72 (1.19)
4.85 (1.70)
3.27 (1.31)
Prideful
4.02 (1.88)
5.35 (1.34)
5.28 (1.21)
3.83 (1.82)
5.26 (1.21)
Impressionable
3.88 (1.69)
4.09 (1.64)
4.65 (1.16)
3.25 (1.53)
4.62 (1.26)
Extravagant
3.77 (1.64)
5.32 (1.72)
4.33 (1.44)
2.76 (1.67)
4.17 (1.46)
Softspoken
3.73 (1.40)
4.76 (1.60)
3.19 (1.31)
4.29 (1.75)
3.00 (1.15)
Cunning
3.69 (1.96)
3.90 (1.80)
3.56 (1.31)
3.13 (1.78)
3.91 (1.21)
Choosy
3.53 (1.41)
5.21 (1.32)
4.75 (1.12)
4.35 (1.75)
4.94 (1.23)
Ordinary
3.53 (1.39)
4.06 (1.69)
5.04 (1.22)
4.23 (1.79)
4.60 (1.54)
Eccentric
3.53 (1.30)
4.16 (1.66)
3.44 (1.35)
3.68 (1.91)
3.36 (1.44)
Strict
3.46 (1.43)
5.40 (1.53)
3.42 (1.22)
3.59 (1.85)
3.50 (1.25)
Self-concerned
3.45 (1.58)
5.11 (1.57)
5.08 (1.37)
4.04 (1.50)
5.14 (1.31)
Daydreamer
3.43 (1.31)
4.30 (1.81)
4.41 (1.38)
4.40 (1.78)
4.12 (1.17)
Solemn
3.37 (1.36)
4.92 (1.53)
3.53 (1.35)
3.63 (1.71)
3.23 (1.05)
Overcautious
3.01 (1.22)
4.69 (1.57)
3.83 (1.22)
4.56 (1.84)
3.37 (1.26)
Inhibited
2.94 (1.29)
4.07 (1.65)
3.52 (1.14)
3.30 (1.55)
3.16 (1.44)
Bashful
2.82 (1.44)
3.76 (1.67)
3.24 (1.33)
3.60 (1.95)
3.00 (1.30)
Melancholy
2.76 (1.62)
4.10 (1.63)
3.51 (1.13)
3.19 (1.88)
3.49 (1.30)

Replication and extension of Alicke (1985) 17
Table 4 (Continued)
Mean ratings and standard deviations of traits in terms of desirability, controllability and
commonness, self-ratings, and other-ratings
Desirability
(N = 149)
Controllability
(N = 149)
Commonness
(N = 149)
Self
(N = 149)
Other
(N = 149)
Traits
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
Irreligious
2.76 (1.29)
5.17 (1.73)
3.51 (1.43)
3.73 (2.32)
3.51 (1.41)
Impulsive
2.75 (1.30)
4.23 (1.71)
4.27 (1.40)
2.81 (1.50)
5.18 (1.26)
Passive
2.73 (1.29)
4.74 (1.43)
3.76 (1.31)
3.55 (1.71)
3.35 (1.39)
Hesitant
2.70 (1.35)
4.52 (1.62)
3.57 (1.14)
4.03 (1.59)
3.52 (1.27)
Meek
2.62 (1.50)
4.18 (1.72)
2.97 (1.30)
3.35 (1.81)
2.88 (1.10)
Compulsive
2.60 (1.40)
3.93 (1.76)
4.05 (1.54)
2.92 (1.69)
4.42 (1.43)
Restless
2.59 (1.35)
4.30 (1.53)
4.01 (1.52)
3.65 (1.54)
4.36 (1.49)
Boastful
2.58 (1.58)
5.52 (1.55)
4.67 (1.33)
2.33 (1.52)
4.88 (1.08)
Radical
2.51 (1.26)
4.92 (1.67)
3.41 (1.52)
2.71 (1.77)
3.18 (1.54)
Timid
2.47 (1.20)
3.72 (1.89)
3.14 (1.36)
3.52 (1.96)
2.99 (1.19)
Profane
2.42 (1.47)
5.57 (1.51)
4.19 (1.45)
2.76 (1.69)
3.84 (1.47)
Unemotional
2.42 (1.34)
3.61 (1.74)
3.04 (1.42)
2.83 (1.84)
2.55 (1.32)
Unpoised
2.26 (1.28)
4.66 (1.52)
3.68 (1.45)
3.00 (1.70)
3.65 (1.40)
Unoriginal
2.23 (1.31)
3.54 (1.55)
3.94 (1.50)
2.80 (1.70)
3.6 (1.61)
Unsophisticated
2.22 (1.27)
4.49 (1.67)
3.94 (1.43)
2.91 (1.62)
4.08 (1.39)
Discontented
2.15 (1.24)
4.93 (1.66)
4.19 (1.51)
2.76 (1.70)
3.95 (1.51)
Self-centered
2.12 (1.21)
5.28 (1.58)
5.00 (1.25)
2.78 (1.53)
4.98 (1.17)
Humorless
2.09 (1.44)
3.62 (1.87)
2.92 (1.27)
1.94 (1.38)
2.90 (1.40)
Uncultured
2.09 (1.25)
4.71 (1.68)
3.70 (1.39)
2.24 (1.32)
3.74 (1.51)
Unstudious
2.08 (1.36)
5.28 (1.70)
3.62 (1.23)
2.32 (1.48)
3.57 (1.30)
Vain
2.08 (1.17)
4.94 (1.63)
4.43 (1.44)
2.48 (1.67)
4.31 (1.50)
Unforgiving
2.07 (1.22)
5.30 (1.51)
3.62 (1.40)
2.68 (1.60)
3.58 (1.55)
Clumsy
2.02 (1.43)
3.52 (1.80)
3.38 (1.34)
3.40 (1.80)
3.17 (1.36)
Forgetful
2.00 (1.17)
3.59 (1.66)
3.81 (1.27)
3.06 (1.63)
3.76 (1.53)
Unentertaining
1.99 (1.39)
4.04 (1.48)
3.48 (1.36)
2.99 (1.70)
3.17 (1.32)
Cold
1.97 (1.20)
5.01 (1.57)
3.43 (1.33)
2.46 (1.50)
3.16 (1.43)
Withdrawn
1.97 (1.17)
4.43 (1.68)
3.01 (1.33)
3.54 (1.93)
2.93 (1.41)
Gullible
1.93 (1.22)
3.76 (1.67)
4.09 (1.33)
2.85 (1.66)
3.90 (1.48)

Replication and extension of Alicke (1985) 18
Table 4 (Continued)
Mean ratings and standard deviations of traits in terms of desirability, controllability and
commonness, self-ratings, and other-ratings
Desirability
(N = 149)
Controllability
(N = 149)
Commonness
(N = 149)
Self
(N = 149)
Other
(N = 149)
Traits
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
Complaining
1.92 (1.41)
5.57 (1.58)
4.65 (1.37)
3.00 (1.61)
4.54 (1.34)
Deceptive
1.91 (1.34)
5.35 (1.68)
3.66 (1.32)
2.17 (1.26)
3.56 (1.42)
Meddlesome
1.91 (1.31)
5.70 (1.35)
3.74 (1.45)
2.21 (1.33)
3.93 (1.28)
Disobedient
1.91 (1.17)
5.82 (1.39)
3.69 (1.49)
2.57 (1.57)
3.69 (1.45)
Maladjusted
1.90 (1.33)
3.86 (1.74)
3.38 (1.48)
2.22 (1.64)
3.17 (1.37)
Dissatisfied
1.90 (1.25)
4.64 (1.68)
4.38 (1.59)
3.14 (1.76)
4.18 (1.51)
Unkind
1.89 (1.32)
5.52 (1.60)
3.30 (1.28)
1.98 (1.32)
3.35 (1.51)
Insecure
1.89 (1.17)
3.75 (1.78)
3.88 (1.55)
3.65 (1.91)
4.01 (1.60)
Irrational
1.87 (1.41)
4.34 (1.74)
3.92 (1.24)
2.26 (1.44)
3.83 (1.31)
Irresponsible
1.85 (1.37)
5.48 (1.57)
3.73 (1.44)
2.32 (1.45)
3.85 (1.42)
Shallow
1.83 (1.14)
4.92 (1.64)
3.99 (1.61)
2.37 (1.59)
4.25 (1.43)
Phony
1.82 (1.41)
5.15 (1.86)
3.58 (1.48)
1.88 (1.37)
3.94 (1.55)
Rude
1.82 (1.36)
5.97 (1.38)
4.00 (1.45)
2.07 (1.45)
3.73 (1.33)
Snobbish
1.81 (1.19)
5.56 (1.61)
3.76 (1.50)
2.14 (1.47)
3.73 (1.53)
Disrespectful
1.80 (1.33)
5.83 (1.46)
3.73 (1.40)
2.02 (1.50)
3.79 (1.59)
Spiteful
1.79 (1.28)
5.14 (1.64)
3.68 (1.46)
2.38 (1.69)
3.6 (1.51)
Uncivil
1.78 (1.28)
5.59 (1.51)
3.22 (1.52)
2.12 (1.61)
3.25 (1.40)
Belligerent
1.78 (1.21)
5.38 (1.52)
3.76 (1.54)
1.98 (1.59)
3.44 (1.52)
Unpopular
1.77 (1.18)
3.46 (1.68)
3.60 (1.35)
3.07 (1.77)
3.44 (1.41)
Unskilled
1.76 (1.16)
4.98 (1.78)
3.64 (1.39)
2.39 (1.51)
3.07 (1.31)
Mean
1.74 (1.17)
5.74 (1.35)
3.53 (1.54)
2.26 (1.62)
3.48 (1.38)
Impolite
1.73 (1.00)
5.98 (1.38)
3.89 (1.42)
2.00 (1.30)
3.73 (1.50)
Unreasonable
1.71 (1.13)
4.76 (1.69)
3.82 (1.36)
1.95 (1.17)
3.63 (1.58)
Tiresome
1.71 (0.99)
4.28 (1.62)
3.67 (1.57)
2.61 (1.70)
3.57 (1.56)
Discourteous
1.70 (1.02)
5.78 (1.56)
3.78 (1.40)
1.90 (1.35)
3.84 (1.49)
Unappreciative
1.66 (1.00)
5.42 (1.75)
4.20 (1.70)
2.04 (1.41)
3.86 (1.48)
Troubled
1.66 (0.95)
3.77 (1.61)
3.92 (1.29)
2.67 (1.65)
3.87 (1.43)
Lazy
1.65 (1.24)
5.59 (1.53)
3.77 (1.43)
2.81 (1.70)
3.78 (1.47)

Replication and extension of Alicke (1985) 19
Table 4 (Continued)
Mean ratings and standard deviations of traits in terms of desirability, controllability and
commonness, self-ratings, and other-ratings
Desirability
(N = 149)
Controllability
(N = 149)
Commonness
(N = 149)
Self
(N = 149)
Other
(N = 149)
Traits
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
Ill-mannered
1.65 (1.18)
5.47 (1.69)
3.62 (1.55)
2.15 (1.63)
3.60 (1.60)
Jealous
1.65 (0.94)
4.57 (1.69)
4.28 (1.41)
2.62 (1.56)
3.79 (1.40)
Unpleasant
1.62 (1.18)
5.24 (1.55)
3.44 (1.30)
2.02 (1.36)
3.34 (1.48)
Hostile
1.59 (1.12)
5.35 (1.40)
3.48 (1.57)
1.84 (1.25)
3.28 (1.53)
Deceitful
1.58 (1.02)
5.51 (1.67)
3.70 (1.38)
1.86 (1.37)
3.16 (1.37)
Unethical
1.57 (1.09)
5.43 (1.54)
3.52 (1.57)
1.65 (1.08)
3.41 (1.49)
Liar
1.57 (1.07)
5.98 (1.44)
3.89 (1.68)
2.05 (1.43)
3.33 (1.36)
Dishonorable
1.55 (1.16)
5.22 (1.89)
3.04 (1.50)
1.90 (1.64)
2.88 (1.51)
Unpleasing
1.54 (0.84)
4.49 (1.65)
3.27 (1.22)
2.37 (1.60)
2.89 (1.31)
Incompetent
1.53 (1.26)
4.40 (1.77)
3.30 (1.33)
1.85 (1.37)
3.17 (1.39)
Dishonest
1.53 (1.08)
5.43 (1.67)
3.42 (1.38)
2.05 (1.37)
3.59 (1.27)
Note. The traits are arranged in descending order of desirability ratings.
Replication
We found strong support for the desirability effect hypothesis. We ran a correlation
analysis and found that desirability had a positive association with self-minus-other ratings (r
= .77, 95% CI [.69, .82], p < .001). This relationship is illustrated in Figure 1. We found
support for differences between the desirability self-ratings correlation (r = .92, 95% CI
[.89, .94], p < .001) and the desirability and other-ratings correlation (r = .61, 95% CI
[.50, .70], p < .001) (comparison: z = 8.76, p < .001).
We conclude that regardless of the rating perspective, participants perceived more
desirable traits as more descriptive of themselves or the average other, and this positive
relationship was stronger for self-ratings than for other-ratings.

Replication and extension of Alicke (1985) 20
Figure 1
Scatterplot with marginal histograms and 95% confidence interval showing the relationship
between desirability and self-minus-other ratings.
We proceeded to conduct a multiple linear regression analysis to investigate whether
desirability and controllability interacted to predict self-minus-other ratings, and summarized
findings in Table 5. First, desirability and controllability were entered into the model. We
found that the overall regression was statistically significant (r = .77, R2 = .59, 95% CI
[.48, .66], F(2, 146) = 104.30, p < .00). Next, the interaction term was added to the model,
which accounted for variance in self-minus-other ratings (ΔR2 = .07, 95% CI [.02, .12], ΔF(1,
145) = 28.17, p < .001). The relationship between desirability and self-minus-other ratings
was moderated by controllability (b = 0.19, SE = 0.04, 95% CI [0.12, 0.26], p < .001). The
interaction was probed by testing the simple main effects of desirability at two levels of
controllability: one standard deviation below the mean and one standard deviation above the

Replication and extension of Alicke (1985) 21
mean. We found that an increase in controllability enhanced the positive relationship between
desirability and self-minus-other ratings (Figure 2). This means that the higher the trait
controllability, the more likely participants were to regard more desirable traits as more
descriptive of themselves than of others.
Figure 2
Simple slopes of desirability predicting self-minus-other ratings for 1 SD below the mean of
controllability and 1 SD above the mean of controllability.

Replication and extension of Alicke (1985) 22
Table 5
Regression results using self-minus-other ratings as the dependent variable, desirability and controllability as the independent variables
Predictor
b
b
95% CI
[LL, UL]
beta
beta
95% CI
[LL, UL]
sr2
sr2
95% CI
[LL, UL]
Fit
Difference
(Intercept)
-0.24***
[-0.36, -0.13]
Desirability
0.46***
[0.40, 0.53]
0.77
[0.66, 0.87]
.59
[.49, .69]
R2 = .59***
Controllability
-0.07
[-0.21, 0.07]
-0.05
[-0.16, 0.05]
.00
[-.01, .01]
95% CI [.48,.66]
(Intercept)
-0.25***
[-0.36, -0.15]
R2 = .66***
ΔR2 = .07***
Desirability
0.45***
[0.39, 0.51]
0.74
[0.64, 0.84]
.54
[.43, .65]
95% CI [.56,.71]
F(3, 145) = 91.84***
95% CI [.02, .12]
ΔF(1, 145) = 28.17***
Controllability
-0.15*
[-0.28, -0.02]
-0.11
[-0.21, -0.01]
.01
[-.01, .03]
Interaction
0.19***
[0.12, 0.26]
0.27
[0.17, 0.37]
.07
[.02, .12]
Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. b represents unstandardized regression
weights. beta indicates the standardized regression weights. sr2 represents the semi-partial correlation squared. r represents the zero-order
correlation. LL and UL indicate the lower and upper limits of a confidence interval, respectively.
* indicates p < .05. ** indicates p < .01. *** indicates p < .001

Replication and extension of Alicke (1985) 23
The effect of desirability was strong, explaining more than half of the variation in
self-minus-other ratings when holding controllability and the interaction term constant (sr2
= .54, 95% CI [.43, .65]). The effect of controllability was not statistically significant when
controlling for other predictors (sr2 = .01, 95% CI [-.01, .03]). The interaction between
desirability and controllability had a greater effect than that of controllability, but noticeably
smaller than that of desirability (sr2 = .07, 95% CI [.02, .12]). Results supported the
hypothesis that desirability and controllability interact to predict the size of the difference
between self-ratings and other-ratings.
Extension: Trait desirability, trait commonness, and self-other ratings
We conducted two multiple linear regression analyses to examine the interaction
between self-ratings and desirability, as well as the interaction between other-ratings and
desirability on predicting commonness. We summarized the results in Tables S7 and S8 in
the supplementary, respectively. We failed to find support for the relationship between
commonness and self or other-ratings being dependent on desirability. Our discussion below
therefore focuses on the first step of each model when only the two predictors were entered.
Our findings did not fully support the extension hypothesis, as we found support for a
relationship between trait desirability and trait commonness, yet found no support for a
relationship between trait desirability and self-ratings.
Examining desirability and self-ratings, we found support for desirability as predictive
of commonness (b = 0.16, SE = 0.05, 95% CI [0.07, 0.26], p < .001), but not for self-ratings
(b = 0.05, SE = 0.07, 95% CI [-0.09, 0.18], p = .48) (see Table S7).
On the other hand, examining desirability and other-ratings, they were both positively
associated with commonness. Other-ratings (b = 0.80, SE = 0.04, 95% CI [0.71, 0.86], p

Replication and extension of Alicke (1985) 24
<.001) was a stronger predictor than desirability (b = 0.04, SE = 0.01, 95% CI [0.01, 0.06], p
= .001) (see Table S8).
Overall, both regression equations accounted for a significant portion of variance in
commonness ratings: R2 = .86, 95% CI [.82, .89], F(2, 146) = 51.88, p < .001 for desirability
and other-ratings and R2 = .42, 95% CI [.29, .51], F(2, 146) = 451, p < .001 for desirability
and self-ratings respectively.
The association between other-ratings and commonness was the strongest (sr2 = .41,
95% CI [.31, .52]), compared with desirability and the interaction term which accounted for
very little variance in commonness in the same model. Desirability was a stronger predictor
of commonness (sr2 = .04, 95% CI [-.01, .09]), than self-ratings and the interaction between
self-ratings and commonness, which showed very weak effects when entered in the same
model.
In our pre-registration, we planned to evaluate replication outcomes based on the
direction and strength of the detected signals in relation to the original effect size at a 95%
confidence interval (LeBel et al., 2018; see Table 6 for a comparison). Given the difference
in our statistical analyses from the original, our findings addressed only some of the effects in
a different approach and may not be applicable for a direct comparison using LeBel et al.’s
(2018) framework. We recommend caution in comparing the effect sizes of Alicke (1985)
and of the present replication.

Replication and extension of Alicke (1985) 25
Table 6
Comparison of effect sizes between the original article and replication, with the self- minus-other ratings difference as dependent variable
Original Article
Replication
NHST Summary
Replication Summary
ηp2
f
B
beta
sr2
Desirability
.78
[.73, .81]
1.88
[1.66, 2.06]
0.45
[0.39, 0.51]
0.74
[.64, .38]
.54
[.43, .65]
Supported
Consistent in direction;
strong effect
Controllability
.06
[.002, .18]
.26
[0.04, 0.47]
-0.15
[-0.28, -0.02]
-0.11
[-0.21, -0.03]
.01
[-.01, .03]
Supported
Inconsistent in direction;
inconclusive finding
Desirability  controllability
.21
[.12, .28]
.52
[ 0.37, 0.62]
0.19
[0.12, 0.26]
0.27
[0.17, 0.37]
.07
[.02, .12]
Supported
Consistent in direction;
weak effect
Desirability  controllability*
.15
[.04, .28]
.42
[0.2, 0.62]
Desirability  perspective
.59
[.52, .65]
1.21
[1.04, 1.35]
Perspective  desirability 
controllability
.23
[.14, .31]
.55
[0.40, 0.66]
Note. Values in square brackets indicate the 95% confidence interval. b represents unstandardized regression weights. beta indicates the
standardized regression weights. sr2 represents the semi-partial correlation squared. * indicates that the original article revised categorization of
desirability at high, neutral-high, neutral-low, low level. Since this analysis is performed with the self- minus-other ratings as dependent
variable, we did not include effects of the desirability x perspective and the perspective  desirability  controllability interactions.

Replication and extension of Alicke (1985) 26
Discussion
The present study aimed to replicate and extend the findings of Alicke’s (1985) study.
Alicke (1985) found support for the effects of trait dimensions on the difference between
self-ratings and other-ratings. In two pre-registered data collections, we successfully
replicated the effects for desirability, as well as the interaction between desirability and
controllability. First, there was a strong positive relationship between trait desirability and the
difference between self-ratings and other-ratings on the same trait. The more participants
rated a trait as desirable, the more participants rated the trait as more characteristic of
themselves than characteristic of the average American. Second, the effect of desirability on
the difference between self-ratings and other-ratings was stronger for highly controllable
traits, and weaker (but still positive) for less controllable traits. However, the main effect of
controllability was found to be weaker than expected, and in the opposite direction to the
original (which was significant and positive). Additionally, in our extension we found that
more desirable traits were regarded as more common, yet this only applied to other ratings,
but not self-ratings.
Replication: Effect of desirability, controllability, and their interaction on Better-than-
Average effect
In summary, the predictors showed similar relative magnitudes as the original study:
desirability showed the strongest effect, followed by the interaction between desirability and
controllability, and then controllability. Of particular interest is the consistently strong effect
observed in desirability across all analyses in the present study, which appears to support the
robustness of the better-than-average effect. Taking into account different trait dimensions,
previous studies found that people tended to rate traits as more characteristic of themselves
than of the average other when the traits were more important or more positive (Brown,
2012; Pahl & Eiser, 2005; Pedregon et al., 2012). A meta-analysis study suggested that

Replication and extension of Alicke (1985) 27
Westerners self-enhanced more than East Asians and that the better-than-average paradigm
yielded one of the strongest effects for self-enhancement in both cultures among 31 methods
(Heine & Hamamura, 2007). This finding, however, has been contradicted by subsequent
research arguing that there is little difference between Westerners and East Asians in the
extent of self-enhancement (Brown, 2010; Zell, Strickhouser, Sedikides, & Alicke, 2019; also
see Ziano et al., 2020, and Chandrasekar et al., 2020 for high consistency in findings across
American and Hong Kong samples in judgement and decision-making effects).
We replicated the interaction between desirability and controllability. This is in
support of related research showing similar moderating effects for controllability, such that
controllable traits were regarded as more self-descriptive when described positively, but less
so when described negatively (Rothermund et al., 2005). On the other hand, we did not
replicate the main effect of controllability, finding a weak and inconsistent effect, with
confidence intervals including the null and the effect in the direction opposite to the original.
Given the deviations in both the magnitude and the direction, we consider this finding
inconclusive.
Extension: Does commonness confound Better-than-Average effects?
We found that desirability was positively associated with commonness, yet we found
no evidence that this relationship moderated the self-other ratings difference.
The positive association between desirability and commonness was supported by
another study by Pahl and Eiser (2005), although authors did not specify how commonness
was operationalized. A possible interpretation for this finding is that desirability might be
confounded with commonness, as argued by Moore (2007). Specifically, social norms not
only determine which traits are desirable but also encourage people to display these traits
more often than to display the undesirable ones. As a result, these traits would be regarded as
more common.

Replication and extension of Alicke (1985) 28
In comparison with self-ratings, other ratings were a stronger predictor of
commonness. Revisiting the definition of commonness may help explain this finding. In the
present study, commonness was operationalized as the degree to which people perceive that a
trait is frequently displayed among the average American. Implicated in this definition is
observability, which may help explain why self-ratings and other ratings correlated with
commonness to varying degrees. For self-evaluation, people can access their inner thoughts
and feelings, and recall different instances when they display a certain trait. However,
evaluating the average American is likely different. Compared with self-evaluation, not only
is it more limited to observable traits, but it also requires additional cognitive effort of
imagining an abstracted average (Chambers & Windschitl, 2004; Krizan & Suls, 2008). In
other words, whereas people are aware of both their public and private traits when making
self-evaluation, they are likely to base their judgment of the average other on observable
traits.
Constraints on generality
Sample. We recruited U.S. residents from MTurk as participants. This limits the
generalizability of the present results to other populations, especially non-WEIRD ones
(Western, Educated, Industrialized, Rich, Democratic; Henrich, Heine, & Norenzayan, 2010).
This is of particular importance given the ongoing controversy in the literature on whether
East Asians and Westerners differ in the extent of self-enhancement (Heine & Hamamura,
2007; Brown, 2010; Zell et al., 2020). Research using a different sample may obtain different
results.
Materials. We had no access to the original list of traits used for the first sample. The
original article mentioned where and how the traits were initially derived, yet the full list was
unreported, leading to our decision to use only the 149 traits provided in the original's
appendix. It is possible that the perception towards the traits has changed over the past three

Replication and extension of Alicke (1985) 29
decades, and thus different traits would have been shortlisted for the second sample based on
the ratings of desirability and controllability. This gap in information calls for more shared
documentation in psychological research for facilitating reproducible work. Research using a
different list of traits may obtain different results.
“Average American” designation. Note that the designation of “average American”
is potentially confusing, as “American” can indicate people who are not U.S. citizens (e.g.,
Bolivians, Mexicans, and Canadians). We used the “average American” designation because
it is used in the U.S. media and in popular culture (e.g., Corley, 2018; O’Keefe, 2012).
Nonetheless, future research should adopt “average U.S. American” when U.S. citizens are
recruited as participants in order to avoid lumping together different nationalities and
cultures, and it may find different results.
Further, the MTurk population may differ from the general U.S. population (e.g.,
some found MTurk workers show lower religiosity, higher education, and lower income;
Clifford, Jewell, & Waggoner, 2015). Therefore, when rating themselves in comparison with
the “average American”, MTurk workers may not be showing bias if they are rating
themselves as below-average in religiosity, but in fact produce an accurate estimation of
comparative religiosity. Future research employing MTurk samples may consider the
implications of using designation of “average MTurk worker” compared to "average
American") in aim of addressing this potential confound.
Conclusion
We successfully replicated Alicke (1985). More than thirty years after the original
finding, people still believe they are better than average on desirable traits. The effect of
desirability on the better-than-average effect is stronger for traits considered controllable.

Replication and extension of Alicke (1985) 30
References
Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and
interpreting interactions. Sage.
Alicke, M. D. (1985). Global self-evaluation as determined by the desirability and
controllability of trait adjectives. Journal of Personality and social Psychology, 49(6),
1621.
Alicke, M. D., & Govorun, O. (2005). The better-than-average effect. The self in social
judgment, 1, 85-106.
Alicke, M. D., Zell, E., & Guenther, C. L. (2013). Social self-analysis: Constructing,
protecting, and enhancing the self. In Advances in Experimental Social Psychology
(Vol. 48, pp. 173-234). Academic Press.
Anderson, N. H. (1968). Likableness ratings of 555 personality-trait words. Journal of
Personality and Social Psychology, 9(3), 272.
Brown, J. D. (2010). Across the (Not So) Great Divide: Cultural Similarities in Self-
Evaluative Processes. Social and Personality Psychology Compass, 4(5), 318–330.
https://doi.org/10.1111/j.1751-9004.2010.00267.x
Brown, J. D. (2012). Understanding the better than average effect: Motives (still) matter.
Personality and Social Psychology Bulletin, 38(2), 209-219.
Brown, J. (2014). The self. Psychology Press.
Chambers, J. R., & Windschitl, P. D. (2004). Biases in social comparative judgments: the
role of nonmotivated factors in above-average and comparative-optimism effects.
Psychological Bulletin, 130(5), 813.
Chandrashekar, S. P., Yeung, S. K., Yau, K. C., Feldman, G., Cheung, C. Y., Agarwal, T.

Replication and extension of Alicke (1985) 31
K.,.. & Li, Y. T. Agency and self-other asymmetries in perceived bias and
shortcomings: Replications of the Bias Blind Spot and extensions linking to free will
beliefs. Retrieved March 2020 from: DOI: 10.13140/RG.2.2.19878.16961/1
Chen, J., Hui, L. S., Yu, T., Feldman, G., Zeng, S. V., Ching, T. L., ... & Cheng, B. L. (2020).
Foregone opportunities and choosing not to act: Replications of Inaction Inertia
effect. Social Psychological and Personality Science. Manuscript accepted for
publication. Retrieved June 2020 from
https://www.researchgate.net/publication/332550110_Foregone_opportunities_and_c
hoosing_not_to_act_Replications_of_Inaction_Inertia_effect
Clifford, S., Jewell, R. M., & Waggoner, P. D. (2015). Are samples drawn from mechanical
turk valid for research on political ideology? Research and Politics, 2(4).
https://doi.org/10.1177/2053168015622072
Collaborative Open-science REsearch (2020). Replications and extensions of classic findings
in Judgment and Decision Making. DOI 10.17605/OSF.IO/5Z4A8. Retrieved June 2020
from http://osf.io/5z4a8 and http://mgto.org/pre-registered-replications/
Coppock, A. (2017). Generalizing from Survey Experiments Conducted on Mechanical Turk:
A Replication Approach. Political Science Research and Methods.
https://doi.org/http://alexandercoppock.com/papers/Coppock_generalizability.pdf
Coppock, A., Leeper, T. J., & Mullinix, K. J. (2018). Generalizability of heterogeneous
treatment effect estimates across samples. Proceedings of the National Academy of
Sciences, 1–15.
Corley, T. (2018). Average in America is a prison—here’s what it looks like and how you
can break free. CNBC.Com. Retrieved from https://www.cnbc.com/2018/02/07/tom-
corley-heres-what-average-looks-like-in-america.html
Ditto, P. H., & Griffin, J. (1993). The value of uniqueness: Self-evaluation and the perceived

Replication and extension of Alicke (1985) 32
prevalence of valenced characteristics. Journal of Social Behavior and
Personality, 8(2), 221.
Dunning, D., Heath, C., & Suls, J. M. (2004). Flawed self-assessment: Implications for
health, education, and the workplace. Psychological Science in the Public
Interest, 5(3), 69-106.
Dunning, D., Meyerowitz, J. A., & Holzberg, A. D. (1989). Ambiguity and self-evaluation:
The role of idiosyncratic trait definitions in self-serving assessments of ability.
Journal of Personality and Social Psychology, 57(6), 1082.
Epley, N., & Dunning, D. (2000). Feeling “holier than thou”: are self-serving assessments
produced by errors in self-or social prediction?. Journal of Personality and Social
Psychology, 79(6), 861.
Frederick, S. (2012). Overestimating Others’ Willingness to Pay. Journal of Consumer
Research, 39(1), 1–21. https://doi.org/10.1086/662060
Guenther, C. L., Taylor, S. G., & Alicke, M. D. (2015). Differential reliance on performance
outliers in athletic self‐assessment. Journal of Applied Social Psychology, 45(7), 374-
382.
Heine, S. J., & Hamamura, T. (2007). In search of East Asian self-enhancement. Personality
and Social Psychology Review, 11(1), 4-27.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? The
Behavioral and Brain Sciences, 33(2–3), 61–83; discussion 83-135.
https://doi.org/10.1017/S0140525X0999152X
Irwin, J. R., & McClelland, G. H. (2003). Negative consequences of dichotomizing
continuous predictor variables. Journal of Marketing Research, 40(3), 366-371.
Kanten, A. B., & Teigen, K. H. (2008). Better than average and better with time: Relative

Replication and extension of Alicke (1985) 33
evaluations of self and others in the past, present, and future. European Journal of
Social Psychology, 38(2), 343-353.
Krizan, Z., & Suls, J. (2008). Losing sight of oneself in the above-average effect: When
egocentrism, focalism, and group diffuseness collide. Journal of Experimental Social
Psychology, 44(4), 929-942.
Krueger, J., & Mueller, R. A. (2002). Unskilled, unaware, or both? The better-than-average
heuristic and statistical regression predict errors in estimates of own performance.
Journal of Personality and Social Psychology, 82(2), 180.
LeBel, E. P., McCarthy, R. J., Earp, B. D., Elson, M., & Vanpaemel, W. (2018). A Unified
Framework to Quantify the Credibility of Scientific Findings. Retrieved April 2018
from http://doi.org/10.17605/OSF.IO/UWMR8 DOI: 10.17605/OSF.IO/UWMR8
Litman, L., Robinson, J., & Abberbock, T. (2017). TurkPrime. Com: A versatile
crowdsourcing data acquisition platform for the behavioral sciences. Behavior
Research Methods, 49(2), 433-442.
Logg, J. M., Haran, U., & Moore, D. A. (2018). Is overconfidence a motivated bias?
Experimental evidence. Journal of Experimental Psychology: General, 147(10),
1445.
Jung, M. H., Moon, A., & Nelson, L. D. (2019). Overestimating the valuations and
preferences of others. Journal of Experimental Psychology: General.
MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D. (2002). On the practice of
dichotomization of quantitative variables. Psychological Methods, 7(1), 19.
McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting interactions and
moderator effects. Psychological Bulletin, 114(2), 376.
Mischel, W., Shoda, Y., & Ayduk, O. (2007). Introduction to personality: Toward an
integrative science of the person. John Wiley & Sons.

Replication and extension of Alicke (1985) 34
Moore, D. A. (2007). Not so above average after all: When people believe they are worse
than average and its implications for theories of bias in social
comparison. Organizational Behavior and Human Decision Processes, 102(1), 42-58.
Mullinix, K. J., Leeper, T. J., Druckman, J. N., & Freese, J. (2015). The Generalizability of
Survey Experiments. Journal of Experimental Political Science, 2(2), 109–138.
https://doi.org/10.1017/XPS.2015.19
O’Keefe, K. (2012). The average American: The extraordinary search for the Nation’s most
ordinary citizen. Lulu Press, Inc
Pahl, S., & Eiser, J. R. (2005). Valence, comparison focus and self-positivity biases: Does it
matter whether people judge positive or negative traits?. Experimental psychology,
52(4), 303-310.
Pedregon, C. A., Farley, R. L., Davis, A., Wood, J. M., & Clark, R. D. (2012). Social
desirability, personality questionnaires, and the “better than average” effect.
Personality and Individual Differences, 52(2), 213-217.
Pronin, E., Lin, D. Y., & Ross, L. (2002). The bias blind spot: Perceptions of bias in self
versus others. Personality and Social Psychology Bulletin, 28(3), 369–381.
https://doi.org/10.1177/0146167202286008
Rothermund, K., Bak, P. M., & Brandtstädter, J. (2005). Biases in self‐evaluation:
moderating effects of attribute controllability. European Journal of Social
Psychology, 35(2), 281-290.
Sedikides, C., Gaertner, L., & Toguchi, Y. (2003). Pancultural self-enhancement. Journal of
Personality and Social Psychology, 84(1), 60.
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive psychology:
Undisclosed flexibility in data collection and analysis allows presenting anything as
significant. Psychological Science, 22(11), 1359-1366.

Replication and extension of Alicke (1985) 35
Stanley, M. L., Henne, P., Iyengar, V., Sinnott-Armstrong, W., & De Brigard, F. (2017). I’m
not the person I used to be: The self and autobiographical memories of immoral
actions. Journal of Experimental Psychology: General, 146(6), 884.
Svenson, O. (1981). Are we all less risky and more skillful than our fellow drivers?. Acta
Psychologica, 47(2), 143-148.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: a social psychological
perspective on mental health. Psychological Bulletin, 103(2), 193.
Malmendier, U., & Tate, G. (2005). CEO overconfidence and corporate investment. The
Journal of Finance, 60(6), 2661-2700.
Zell, E., Strickhouser, J. E., Sedikides, C., & Alicke, M. D. (2019). The Better-Than-Average
Effect in Comparative Self- Evaluation: A Comprehensive Review and Meta-
Analysis. Psychological Bulletin, 146(2), 118–149.
https://doi.org/10.1037/bul0000218
Ziano, I., & Villanova, D. (2020). More Useful to You: Overestimating Products’Usefulness
to Others Because Of Self-Serving Materialism Attributions. Retrieved from
https://psyarxiv.com/938m7/
Ziano, I., Wang, Y. J., Sany, S. S., Feldman, G., Ngai, L. H., Lau, Y. K., ... & Cheng,
B. L. (2020). Perceived morality of direct versus indirect harm: Replications of the
preference for indirect harm effect. Manuscript accepted for publication.
https://psyarxiv.com/bs7jf DOI: 10.31234/osf.io/bs7jf
Zwaan, R. A., Etz, A., Lucas, R. E., & Donnellan, M. B. (2018). Making replication
mainstream. Behavioral and Brain Sciences, 41.

Alicke (1985) replication and extension:
Supplementary
Table of contents
Disclosures ........................................................................................................................... 3
Procedure ...................................................................................................................................... 3
Pre-registration .............................................................................................................................. 3
Data Exclusion .............................................................................................................................. 3
Conditions Reporting .................................................................................................................... 3
Variables Reporting....................................................................................................................... 3
Attention checks and exclusion checks ............................................................................... 4
Attention checks. 1st data collection ............................................................................................... 4
Attention checks. 2nd data collection .............................................................................................. 4
Exclusion criteria .......................................................................................................................... 5
Tables and figures ............................................................................................................... 6
Table S1. Reported statistics and calculated effect sizes in the original study ................................. 6
Figure S1. Mean pre-ratings of traits in first-wave sample (Alicke, 1985, pp. 1629-1630). ............. 7
Table S2: Ratings in second-wave sample by levels of desirability and controllability in the original
study ............................................................................................................................................. 9
Table S3. Summary of effect sizes using commonness as the dependent variable ......................... 10
Table S4. Comparison of study characteristics between the original article and the replication ..... 11
Table S5. Summary of study design ............................................................................................. 12
Table S7. Regression results using commonness as the dependent variable, desirability and self-
ratings as the independent variables ............................................................................................. 13
Table S8. Regression results using commonness as the dependent variable, desirability and other-
ratings as the independent variables ............................................................................................. 14
Table S9. Simple main effects of desirability on self-minus-other ratings ..................................... 15
Table S10. Mean pre-ratings of the revised conditions ................................................................. 16
Comparison with the Original Article .............................................................................. 17
Table S11. Similarities and differences between the original article and replication study in the
first-wave sample ........................................................................................................................ 17
Table S12. Similarities and differences between the original article and replication study in the
second-wave sample .................................................................................................................... 18
Pre-registration Planning and Deviation Documentation ............................................... 19
Table S13. Pre-registration planning and deviation documentation............................................... 19
Materials ............................................................................................................................ 20
Qualtrics Surveys ........................................................................................................................ 20
Rating Criteria............................................................................................................................. 20

Replication and extension of Alicke (1985): Supplementary 2
List of Traits for Ratings ............................................................................................................. 21
Effect Sizes and Confidence Intervals .............................................................................. 22
Power Analyses.................................................................................................................. 29
Statistical Assumptions and Normality Tests ................................................................... 30
Before Exclusion ......................................................................................................................... 30
Figure S2. Residuals versus fitted plot for self-minus-other ratings predicted from desirability and
controllability before exclusion. .................................................................................................. 30
Figure S3. Normal Q-Q plot for self-minus-other ratings predicted from desirability and
controllability before exclusion. .................................................................................................. 31
Figure S4. Residuals versus fitted plot for commonness predicted from desirability and self-ratings
before exclusion .......................................................................................................................... 32
Figure S5. Normal Q-Q plot for commonness predicted from desirability and self-ratings before
exclusion. .................................................................................................................................... 33
Figure S6. Residuals versus fitted plot for commonness predicted from desirability and other
ratings before exclusion. .............................................................................................................. 34
Figure S7. Normal Q-Q plot for commonness predicted from desirability and other ratings before
exclusion. .................................................................................................................................... 35
After Exclusion ........................................................................................................................... 36
Figure S8. Residuals versus fitted plot for self-minus-other ratings predicted from desirability and
controllability after exclusion. ..................................................................................................... 36
Figure S9. Normal Q-Q plot for self-minus-other ratings predicted from desirability and
controllability after exclusion ...................................................................................................... 37
Figure S10. Residuals versus fitted plot for commonness predicted from desirability and self-
ratings after exclusion. ................................................................................................................ 38
Figure S11. Normal Q-Q plot for commonness predicted from desirability and self-ratings after
exclusion. .................................................................................................................................... 39
Figure S12. Residuals versus fitted plot for commonness predicted from desirability and other
ratings after exclusion. ................................................................................................................ 40
Figure S13. Normal Q-Q plot for commonness predicted from desirability and other ratings after
exclusion. .................................................................................................................................... 41
Exploratory Analyses ........................................................................................................ 42
Table S14. Summary of correlation comparisons and effect sizes ................................................. 42
Results after Exclusion ...................................................................................................... 43
Table S15. Summary of demographics of the first-wave and second-wave samples after exclusion
................................................................................................................................................... 43
Table S16. Means, standard deviations, and correlations with confidence intervals after exclusion
................................................................................................................................................... 44
Figure S14. Scatterplot showing the relationship between desirability and self-minus-other ratings
with 95% confidence interval after exclusion. .............................................................................. 45
Table S18. Regression results using commonness as the criterion after exclusion ......................... 47
Table S19. Regression results using commonness as the criterion after exclusion ......................... 48

Replication and extension of Alicke (1985): Supplementary 3
Disclosures
Procedure
The replication was conducted as part of a large replication project, in which we attempted to
replicate findings from the judgment and decision-making literature. In the present study, the
participants from both the initial and second-wave samples received financial compensation
for completing a survey.
Pre-registration
We pre-registered the study prior to data collection. The design and analysis plan were
revised after our analysis of the first data collection.
All departures from the pre-registration are documented in the manuscript or in the
supplementary below, see section "Pre-registration Planning and Deviation Documentation".
Data Exclusion
We pre-registered exclusion criteria such as low English proficiency and failed
comprehension checks. We conducted our analyses both with and without exclusions, and
found that exclusions had little effect on the results. For the sake of brevity, the manuscript
reported findings without data exclusion.
Conditions Reporting
All conditions collected for this study are reported and included in the provided data.
Variables Reporting
All variables collected for this study are reported and included in the provided data.

Replication and extension of Alicke (1985): Supplementary 4
Attention checks, Comprehension checks, and Exclusion criteria
Attention checks. 1st data collection
To test whether participants answered the questions carefully, we added two attention
checks to each condition, mixed with the trait ratings. These items were “very undesirable”
and “very desirable” in the desirability condition, and “very uncontrollable” and “very
controllable” in the controllability condition respectively. Participants needed to rate each of
the attention check traits with the option that corresponded exactly to the trait in order to pass
the attention checks. This means that participants passed the attention check if they rated
“very undesirable” or “very uncontrollable” as 1 (corresponding to ““very undesirable” in the
desirability condition or “very uncontrollable” in the controllability condition) and “very
desirable” or “very controllable” as 7 (corresponding to “very desirable” in the desirability
condition or to “very controllable” in the controllability condition).
Attention checks. 2nd data collection
Similar to the first data collection, we added two attention checks for each condition,
mixed with the trait ratings. These items were “very uncommon” and “very common” in the
commonness condition, and “not at all characteristic” and “very characteristic” in the self-
ratings condition respectively. Similar to the first data collection, participants were supposed
to check the option “very common” for the trait “very common”, very uncommon for the trait
“very uncommon”, “not at all characteristic” for the trait “not at all characteristic”, and “very
characteristic” for the trait “characteristic”. Originally, we planned to use “not at all
characteristic” and “very characteristic” in the other ratings condition. However, there was a
coding error in the Qualtrics survey, which rendered the attention checks for this condition
ineffective. This error did not impact the results reported in the following section since we
pre-registered the use of a full sample for analyses. Details are reported in “Pre-registration
Planning and Deviation Document” in the supplementary materials.
Comprehension questions.1st data collection.
At the beginning of the first survey, participants received instructions about the rating
criteria specific to their assigned condition: desirability or controllability. To test participants’
understanding of the rating criteria, the instructions were followed by three comprehension
questions with three multiple choices each. Participants had to answer all comprehension
questions correctly in order to proceed to the rating task.
In the desirability condition, participants were first asked whether a desirable
characteristic or trait is one that is good to have, bad to have, or neither good or bad. Second,
they were asked whether an undesirable characteristic or trait is one that is good to have, bad
to have, or neither good or bad. Third, they were asked whether the task is to make
evaluations based on their own desirability criteria, desirability criteria for the average
American, or whatever desirability criteria seem relevant.
In the controllability condition, participants were first asked whether a controllable
characteristic or trait is one that a person could create or eliminate through sufficient effort,
or a person’s effort would not be sufficient to create or eliminate, or unrelated to persons.
Second, they were asked whether an uncontrollable characteristic or trait is one that a person
could create or eliminate through sufficient effort, or a person’s effort would not be sufficient
to create or eliminate, or unrelated to persons. Third, they were asked whether the task is to
make evaluations based on their own controllability criteria, controllability criteria for the
average American, or whatever controllability criteria seems relevant.

Replication and extension of Alicke (1985): Supplementary 5
Comprehension questions. 2nd data collection.
At the beginning of the second survey, participants received instructions about the
rating criteria specific to their assigned condition: commonness, self-ratings, or other ratings.
Similar to the structure of the first survey, the instructions were followed by one to three
comprehension questions with three multiple choices each. Participants had to answer all
comprehension questions correctly in order to proceed to the rating task.
In the commonness condition, participants were first asked whether a common
characteristic or trait is one that is frequently displayed, rarely displayed, or neither
frequently nor rarely displayed. Second, they were asked the same question for an uncommon
characteristic or trait. Third, they were asked whether the task is to make evaluations based
on their own commonness criteria, commonness criteria for the average American, or
whatever commonness criteria seems relevant.
In the self-ratings condition, they were asked whether the evaluation is based on how
well the traits characterize them, the average American, or everyone. The other ratings
condition comprised the same comprehension question as the self-ratings condition.
Exclusion criteria
As pre-registered, the analyses focused on the full sample. For supplementary
analysis, the following exclusion criteria were pre-registered: (1) participants who reported
low English proficiency (lower than 5 on a scale of 1 to 7); (2) those who reported not being
serious about filling in the survey (lower than 4 on a scale of 1 to 7); (3) those who correctly
guessed the study hypothesis in the funneling section; (4) those who failed to complete the
survey; (5) those who failed to pass the attention check; and (6) those who completed the
survey within less than one minute. Exclusion had little to no effects on results, and analyses
including only participants fulfilling the preregistered criteria are reported in the “Results
after Exclusions” section of these Supplementary materials.

Replication and extension of Alicke (1985): Supplementary 6
Tables and figures
Table S1. Reported statistics and calculated effect sizes in the original study
Reported statistics
Calculated effect sizes
F
df
p
ηp2
f
Desirability
306.80
3, 261
< .0001
.78
[.73, .81]
1.88
[1.66, 2.06]
Controllability
5.93
1, 87
< .02
.06
[.002, .18]
.26
[0.04, 0.47]
Desirability  controllability
22.72
3, 261
< .0001
.21
[.12, .28]
.52
[0.37, 0.62]
Desirability  controllability
14.87
1, 87
< .0005
.15
[.04, .28]
.42
[0.2, 0.62]
Desirability  perspective
126.74
3, 261
< .0001
.59
[.52, .65]
1.21
[1.04, 1.35]
Perspective  desirability 
controllability
25.90
3, 261
< .0001
.23
[.14, .31]
.55
[0.40, 0.66]
Note. Values in square brackets indicate the 95% confidence interval for each effect size. df
indicates degrees of freedom. ηp2 indicates partial eta squared. f indicates Cohen’s f.
Calculations can be found in “Effect Sizes and Confidence Intervals” in the supplementary
materials.

Replication and extension of Alicke (1985): Supplementary 7
Figure S1. Mean pre-ratings of traits in first-wave sample (Alicke, 1985, pp. 1629-1630).

Replication and extension of Alicke (1985): Supplementary 8

Replication and extension of Alicke (1985): Supplementary 9
Table S2: Ratings in second-wave sample by levels of desirability and controllability in the
original study
Level of desirability
Level of control
High
Moderate-high
Moderate-low
Low
Ratings of self
High
5.72 (0.57)
4.60 (0.79)
3.40 (0.73)
2.23 (0.73)
Low
5.37 (0.66)
4.60 (0.54)
3.21 (0.74)
2.59 (0.69)
Ratings of average college student
High
4.69 (0.72)
4.44 (0.72)
3.74 (0.61)
3.26 (0.83)
Low
4.87 (0.74)
4.27 (0.47)
3.40 (0.55)
3.40 (0.78)
Ratings of self minus average college student
High
1.03
0.16
-0.34
-1.03
Low
0.50
0.33
-0.19
-0.81
Note. Values in parentheses are standard deviations.

Replication and extension of Alicke (1985): Supplementary 10
Table S3. Summary of effect sizes using commonness as the dependent variable
Effect sizes
NHST Summary
b
beta
sr2
Desirability
0.16
[0.06, 0.26]
0.53
[0.21, 0.85]
.04
[-.01, .09]
Supported
Self-ratings
0.05
[-0.09, 0.18]
0.11
[-0.21, 0.43]
.00
[-.01, .01]
Not supported
Desirability  self-ratings
0.01
[-0.03, 0.05]
0.03
[-0.10, 0.16]
.00
[-.01, .01]
Not supported
Desirability
0.04
[0.01, 0.06]
0.12
[0.04, 0.20]
.01
[-.00, .02]
Supported
Other ratings
0.80
[0.72, 0.87]
0.86
[0.77, 0.94]
.41
[.31, .52]
Supported
Desirability  other ratings
0.02
[-0.02, 0.07]
0.03
[-0.03, 0.10]
.00
[-.00, .00]
Not supported
Note. b represents unstandardized regression weights. beta indicates the standardized
regression weights. sr2 represents the semi-partial correlation squared. NHST represents null
hypothesis significance testing. NHST summary concerns the main effects and interactions of
the following extension hypothesis: for ratings of others, trait desirability is positively
associated with trait commonness. For ratings of self, trait desirability is negatively
associated with trait commonness.

Replication and extension of Alicke (1985): Supplementary 11
Table S1. Comparison of study characteristics between the original article and the replication
Study
Alicke (1985)
Replication
Sample
Initial
Final
Initial
Final
n
80 (desirability)
/ 84
(controllability)
88 (self) /
88 (other)
341 (desirability)
/ 329
(controllability)
300 (self) /
306 (other) /
297
(commonness)
% Female
57.9
58.0
47.2
54.4
Age M
(Years)
Unreported
Unreported
39.12
39.34
Age SD
(Years)
Unreported
Unreported
12.01
12.42

Replication and extension of Alicke (1985): Supplementary 12
Table S2. Summary of study design
Hypothesis 1 (Replication)
IV 1:
Desirability
IV 2 Condition 1:
Self-perspective
DV:
Title: Self-minus-other ratings of the traits
Specific DV item: Rate to which degree each trait characterizes
you/the average American on a 7-point scale (1 = not at all
characteristic; 7 = very characteristic).
IV 2 Condition 2:
Other perspective
Hypothesis 2 (Replication)
IV 1:
Desirability
IV 2:
Controllability
IV 3 Condition 1:
Self-perspective
DV:
Title: Self-minus-other ratings of the traits
Specific DV item: Rate to which degree each trait characterizes
you/the average American on a 7-point scale (1 = not at all
characteristic; 7 = very characteristic).
IV 3 Condition 2:
Other perspective
Hypothesis 3 (Extension)
IV 1:
Desirability
IV 2 Condition 1:
Self-perspective
DV:
Title: Commonness ratings of the traits
Specific DV item: Rate to which degree each trait is common
among the average Americans on a 7-point scale (1 = not at all
common; 7 = very common).
IV 2 Condition 2:
Other perspective
Note. IV represents independent variable. DV represents dependent variable. In the present
study, self-minus-other ratings were calculated by subtracting other ratings from other ratings
to account for the rating perspective.

Replication and extension of Alicke (1985): Supplementary 13
Table S7. Regression results using commonness as the dependent variable, desirability and self-ratings as the independent variables
Predictor
b
b
95% CI
[LL, UL]
beta
beta
95% CI
[LL, UL]
sr2
sr2
95% CI
[LL, UL]
Fit
Difference
(Intercept)
4.07***
[4.00, 4.13]
R2 = .42***
Desirability
0.16**
[0.07, 0.26]
0.54
[0.22, 0.86]
.04
[-.01, .10]
95% CI [.29,.51]
F(2, 146) = 51.88***
Self-ratings
0.05
[-0.09, 0.18]
0.11
[-0.21, 0.43]
.00
[-.01, .01]
(Intercept)
4.05***
[3.93, 4.16]
Desirability
0.16**
[0.06, 0.26]
0.53
[0.21, 0.85]
.04
[-.01, .09]
R2 = .42***
ΔR2 = .001
Self-ratings
0.05
[-0.09, 0.18]
0.11
[-0.21, 0.43]
.00
[-.01, .01]
95% CI [.29, .51]
F(3, 145) = 34.48***
95% CI [-.01, .01]
ΔF(1, 145) = 0.22
Interaction
0.01
[-0.03, 0.05]
0.03
[-0.10, 0.16]
.00
[-.01, .01]
Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. b represents unstandardized regression
weights. beta indicates the standardized regression weights. sr2 represents the semi-partial correlation squared. r represents the zero-order
correlation. LL and UL indicate the lower and upper limits of a confidence interval, respectively.
* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

Replication and extension of Alicke (1985): Supplementary 14
Table S8. Regression results using commonness as the dependent variable, desirability and other-ratings as the independent variables
Predictor
b
b
95% CI
[LL, UL]
beta
beta
95% CI
[LL, UL]
sr2
sr2
95% CI
[LL, UL]
Fit
Difference
(Intercept)
4.07***
[4.03, 4.10]
R2 = .86***
Desirability
0.04**
[0.02, 0.06]
0.13
[0.05, 0.21]
.01
[-.00, .02]
95% CI [.82,.89]
Other ratings
0.79***
[0.71, 0.86]
0.84
[0.77, 0.92]
.45
[.34, .56]
F(2, 146) = 451***
(Intercept)
4.05***
[4.01, 4.10]
R2 = .86***
ΔR2 = .001
Desirability
0.04**
[0.01, 0.06]
0.12
[0.04, 0.20]
.01
[-.00, .02]
95% CI [.82,.89]
F(3, 145) = 300.8***
95% CI [-.00, .00]
ΔF(1, 145) = 0.92
Other ratings
0.80***
[0.72, 0.87]
0.86
[0.77, 0.94]
.41
[.31, .52]
Interaction
0.02
[-0.02, 0.07]
0.03
[-0.03, 0.10]
.00
[-.00, .00]
Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. b represents unstandardized regression
weights. beta indicates the standardized regression weights. sr2 represents the semi-partial correlation squared. r represents the zero-order
correlation. LL and UL indicate the lower and upper limits of a confidence interval, respectively.
* indicates p < .05. ** indicates p < .01. *** indicates p <. 001

Replication and extension of Alicke (1985): Supplementary 15
Table S9. Simple main effects of desirability on self-minus-other ratings
Controllability
b
p
95% CI
One SD below mean***
0.29
< .001
[.21, .38]
One SD above mean***
0.60
< .001
[.52, .68]
Note. *** indicates p < .001. b represents unstandardized regression weights. 95% CI
represents 95% confidence interval.

Replication and extension of Alicke (1985): Supplementary 16
Table S10. Mean pre-ratings of the revised conditions
Revised Conditions
Mean Pre-rating
Mean Difference
Neutral-high D, high C
4.84
0.53
Neutral-high D, low C
5.06
0.32
Neutral-low D, high C
3.89
-0.17
Neutral-low D, low C
3.88
0.22
Note. D refers to desirability. C refers to controllability.

Replication and extension of Alicke (1985): Supplementary 17
Comparison with the Original Article
The below tables summarize and explain the similarities and differences between the original
article and replication study.
Table S11. Similarities and differences between the original article and replication study in
the first-wave sample
Item
Explanation
Original Article
Replication Study
Instructions
Participants in the first-wave sample were
asked them to judge to what extent the
traits were desirable or controllable.
Definitions of desirable and controllable
were given (see Procedures in Section 3 for
details).
Same instructions
Measures/
Stimulus
362 traits
149 traits (The article reported using 154
traits (Alicke, 1985, p. 1624) but the
appendix listed only 149 traits.)
Paper-and-pencil survey
● One booklet (either desirability or
controllability)
● Sheets in randomized order (37 traits
on each)
● Non-randomized choices
Online Qualtrics survey
● Randomized, evenly presented blocks
for desirability and controllability
● 36-40 traits in total
● Added 2 attention checks for each
condition
● Added 3 comprehension questions for
each condition
7-point bipolar scale (1 = not at all
desirable or controllable, 7 = very
characteristic of desirable or controllable)
Same scale
Procedure
Between-subjects design
Same design
Participants rated all traits on either
desirability or controllability
Participants were randomly assigned to rate
40 traits either desirability or
controllability
Location
In groups (18 to 29 subjects); location
unreported
Alone; online
Remuneration
Unreported
Participants received 0.5 USD for a task
estimated at 4 minutes, which is
commensurate with the federal minimum
hourly wage of 7.25 USD.
Participant
Population
Introductory psychology students at
University of North Carolina at Chapel
Hill, North Carolina
Americans recruited via Amazon
Mechanical Turk (MTurk)

Replication and extension of Alicke (1985): Supplementary 18
Table S12. Similarities and differences between the original article and replication study in
the second-wave sample
Item
Explanation
Original Article
Replication Study
Instructions
Participants in the second-wave sample
received the first booklet of traits for one
perspective and were asked to rate to which
degree the traits characterized them or the
average college student. Then they
received the second booklet and repeated
the same process for the other perspective.
Participants in the second-wave sample
were asked to rate to what extent the traits
were characteristic of either them or the
average American, or to what extent the
traits are common among the average
American.
Measures/
Stimulus
154 traits
149 traits (The article reported using 154
traits (Alicke, 1985, p. 1624) but the
appendix listed only 149 traits.)
Paper-and-pencil survey
● Two booklets (self & average college
student) presented in counterbalanced
order
● Sheets in randomized order (6 traits
on each)
● Non-randomized choices
Online Qualtrics survey
● 3 randomized blocks: self-ratings,
other ratings or commonness ratings
● Added 2 attention checks for each
condition
● Added 3 comprehension questions for
the commonness condition, and 1
comprehension question each for the
self-condition and the other condition
7-point bipolar scale (1 = not at all
characteristic of me or the average college
student, 7 = very characteristic of me or the
average college student)
Same scale but we replaced “average
college student” with “average American”
to match with our target population
Procedure
Within-subjects design
Between-subjects design
Participants rated all traits in both the self
and other conditions
Participants were randomly assigned to rate
40 traits from the self or average American
perspective
Location
In groups (18 to 29 subjects);
location unreported
Alone; online
Remuneration
Unreported
Participants received 0.5 USD for a task
estimated at 4 minutes, which is
commensurate with the federal minimum
hourly wage of 7.25 USD.
Participant
Population
Introductory psychology students at
University of North Carolina at Chapel
Hill, North Carolina
Americans recruited via MTurk

Replication and extension of Alicke (1985): Supplementary 19
Pre-registration Planning and Deviation Documentation
The below table summarizes the components where there were deviations from the pre-registration.
Table S13. Pre-registration planning and deviation documentation
Components in
your
preregistration
Location of
preregistered
decision/plan
Location of the
rationale for the
decision/plan (if any)
Were there
deviations?*
If yes - describe details of
deviation(s)
Rationale for
deviation
How might the results
be different if you had
not deviated
Procedures
Page 12 of pre-
registration
Page 12 of pre-
registration
No
N/A
N/A
N/A
Power analysis
Page 13 of pre-
registration
Page 13 of pre-
registration
No
N/A
N/A
N/A
Exclusion rules
Page 13 of pre-
registration
Page 13 of pre-
registration
Minor
There was an error in Qualtrics,
which rendered the attention
checks ineffective in the “Other
ratings” condition.
Results after
exclusion in
supplementary
material
The size of the second-
wave sample after
exclusion would be
slightly smaller.
Evaluation criteria
Page 16 of pre-
registration
Page 16 of pre-
registration
Minor
Commented on magnitude and
direction only instead of using
LeBel et al.’s (2018) framework
See discussion of the
manuscript
N/A
Analyses
Page 17-19 of pre-
registration
Page 17-20 of pre-
registration
No
N/A
N/A
N/A
Presentation of
statistics
Page 20 of pre-
registration
Page 20 of pre-
registration
Minor
Did not include a graph for the
extension hypothesis
Weak to no
moderating effects
detected
N/A
Note. *Categories for deviations: Minor - Change probably did not affect results or interpretations; Major - Change likely affected results or interpretations.

Replication and extension of Alicke (1985): Supplementary 20
Materials
Qualtrics Surveys
The full surveys, including the survey flow, randomization options and debrief, are available
in .doc and .qsf file types on the OSF (see main manuscript for links).
Rating Criteria
Each participant was shown 40 of the 149 randomized traits (see list at the end of this
section), and asked to rate these traits based on one of the five rating criteria below:
Desirability
For each of the following:
To what extent do these traits represent desirable or undesirable characteristics for the
average American?
In this context, a desirable characteristic is one that the average American would perceive as
being good to have, whereas an undesirable characteristic is one that the average American
would perceive as being bad to have.
(1 = very undesirable; 7 = very desirable)
Controllability
To what extent do these traits represent controllable or uncontrollable characteristics for the
average American?
A controllable characteristic is one that an average American could create or eliminate
through a sufficient amount of effort, whereas an uncontrollable characteristic is one that
an average American's effort would not be sufficient to create or eliminate.
(1 = very uncontrollable; 7 = very controllable)
Commonness
For each of the following:
To what extent are these traits common among the average Americans?
In this context, a common characteristic is one that the average American would frequently
display, whereas an uncommon characteristic is one that the average American would rarely
display.
(1 = very uncommon; 7 = very common)
Self Ratings
For each of the following:
To what extent do these traits characterize you?
(1 = not at all characteristic of me; 7 = very characteristic of me)
Other Ratings
For each of the following:
To what extent do these traits characterize the average American?
(1 = not at all characteristic of the average American; 7 = very characteristic of the average
American)

Replication and extension of Alicke (1985): Supplementary 21
The below is a full list of the traits used for participant ratings. We referenced the traits
reported in the appendix of the original study.
List of Traits for Ratings
1. Cooperative
2. Considerate
3. Responsible
4. Friendly
5. Respectful
6. Reliable
7. Resourceful
8. Polite
9. Dependable
10. Trustful
11. Pleasant
12. Sincere
13. Loyal
14. Self-disciplined
15. Kind
16. Clean
17. Good-tempered
18. Versatile
19. Persistent
20. Well read
21. Sensitive
22. Grateful
23. Thrifty
24. Neat
25. Bold
26. Self-satisfied
27. Religious
28. Self-concerned
29. Radical
30. Obedient
31. Fashionable
32. Prideful
33. Prudent
34. Choosy
35. Troubled
36. Boastful
37. Unpoised
38. Jealous
39. Self-centered
40. Unskilled
41. Melancholy
42. Unsophisticated
43. Clumsy
44. Daydreamer
45. Irreligious
46. Strict
47. Conforming
48. Compulsive
49. Hesitant
50. Eccentric
51. Unforgiving
52. Disobedient
53. Deceptive
54. Disrespectful
55. Snobbish
56. Spiteful
57. Meddlesome
58. Complaining
59. Unstudious
60. Uncivil
61. Unappreciative
62. Unpleasing
63. Phony
64. Discourteous
65. Unkind
66. Rude
67. Impolite
68. Dishonest
69. Cold
70. Dishonorable
71. Deceitful
72. Hostile
73. Irresponsible
74. Unreasonable
75. Creative
76. Bright
77. Imaginative
78. Intelligent
79. Clear-headed
80. Observant
81. Perceptive
82. Level-headed
83. Mature
84. Honorable
85. Lively
86. Clever
87. Admirable
88. Wise
89. Intellectual
90. Sportsmanlike
91. Punctual
92. Original
93. Interesting
94. Humorous
95. Reserved
96. Cunning
97. Fearless
98. Meticulous
99. Impulsive
100. Ordinary
101. Impressionable
102. Authoritative
103. Normal
104. Attractive
105. Lucky
106. Ingenious
107. Changeable
108. Witty
109. Philosophical
110. Ethical
111. Quick
112. Progressive
113. Sharp-witted
114. Forgetful
115. Uncultured
116. Discontented
117. Dissatisfied
118. Withdrawn
119. Unoriginal
120. Tiresome
121. Profane
122. Unentertaining
123. Passive
124. Timid
125. Bashful
126. Restless
127. Unpopular
128. Unemotional
129. Meek
130. Overcautious
131. Inhibited
132. Extravagant
133. Solemn
134. Softspoken
135. Insecure
136. Belligerent
137. Humorless
138. Lazy
139. Vain
140. Gullible
141. Liar
142. Unpleasant
143. Mean
144. Maladjusted
145. Unethical
146. Ill-mannered
147. Incompetent
148. Shallow
149. Irrational

Replication and extension of Alicke (1985): Supplementary 22
Effect Sizes and Confidence Intervals
Confidence intervals for eta-squared in the original article were calculated using the below
software:
● ηp2 calculation: https://effect-size-calculator.herokuapp.com/#partial-eta-squared-
fixed-effects
● ηp2 to f conversion: https://www.psychometrica.de/effect_size.html#transform
For effect size conversions to f, we used eta-squared to six or seven decimal places (as shown
in the screenshots below) for the estimate and the values within the 95% confidence interval.
In the final manuscript, we reported the f values for the effect sizes.
Main effects:
(1) Desirability:
● Reported: F(3, 261) = 306.80, p < .0001
● Calculated effect sizes:
○ ηp2 = .78, 95% CI [.73, .81]
○ f = 1.88, 95% CI [1.66, 2.06]

Replication and extension of Alicke (1985): Supplementary 23
(2) Controllability:
● Reported: F(1, 87) = 5.93, p < .02
● Calculated:
○ ηp2 = .06, 95% CI [.002, .18]
○ f = .26, 95% CI [0.04, 0.47]

Replication and extension of Alicke (1985): Supplementary 24

Replication and extension of Alicke (1985): Supplementary 25
Interactions:
(1) Desirability x controllability (for revised categorisation: high, neutral-high, neutral-low,
low desirability):
● Reported: F(1, 87) = 14.87, p < .0005
● Calculated:
○ ηp2 = 0.15, 95% CI [.04, .28]
○ f = .42, 95% CI [0.19, 0.62]

Replication and extension of Alicke (1985): Supplementary 26
(2) Desirability x controllability:
● Reported: F(3, 261) = 22.72, p < .0001
● Calculated:
○ ηp2 = 0.21, 95% CI [.12, .28]
○ f = .51, 95% CI [0.37, 0.63]

Replication and extension of Alicke (1985): Supplementary 27
(2) Desirability x perspective:
● Reported: F(3, 261) = 126.74, p < .0001
● Calculated:
○ ηp2 = 0.59, 95% CI [.52, .65]
○ f = 1.21, 95% CI [1.04, 1.35]

Replication and extension of Alicke (1985): Supplementary 28
(3) Perspective x desirability x controllability:
● Reported: F(3, 261) = 25.90, p < .0001
● Calculated:
○ ηp2 = .23, 95% CI [.14, .31]
○ f = .55, 95% CI [.40, .66]

Replication and extension of Alicke (1985): Supplementary 29
Power Analyses
Using G*Power Version 3.1.9.3, we conducted the below power analysis to derive a
minimum sample size of 71 participants. Below is the protocol of the power analysis:
F tests - ANOVA: Fixed effects, special, main effects and interactions
Analysis: A priori: Compute required sample size
Input: Effect size f = 0.511
α err prob = 0.05
Power (1-β err prob) = 0.95
Numerator df = 3
Number of groups = 8
Output: Noncentrality parameter λ = 18.5395910
Critical F = 2.7505411
Denominator df = 63
Total sample size = 71
Actual power = 0.9528557
Note. We pasted the incorrect power analysis protocol in an earlier version of a pre-
registration, using ANCOVA. However, it did not affect the final sample size.

Replication and extension of Alicke (1985): Supplementary 30
Statistical Assumptions and Normality Tests
We conducted a series of tests of statistical assumptions for analyses. These tests include: a)
residual analysis (using residuals versus fitted plot) and normality of residuals (using Q-Q
plot). Below are the plots for the results before and after exclusion.
Before Exclusion
Figure S2. Residuals versus fitted plot for self-minus-other ratings predicted from desirability
and controllability before exclusion.

Replication and extension of Alicke (1985): Supplementary 31
Figure S3. Normal Q-Q plot for self-minus-other ratings predicted from desirability and
controllability before exclusion.

Replication and extension of Alicke (1985): Supplementary 32
Figure S4. Residuals versus fitted plot for commonness predicted from desirability and self-
ratings before exclusion
.

Replication and extension of Alicke (1985): Supplementary 33
Figure S5. Normal Q-Q plot for commonness predicted from desirability and self-ratings
before exclusion.

Replication and extension of Alicke (1985): Supplementary 34
Figure S6. Residuals versus fitted plot for commonness predicted from desirability and other
ratings before exclusion.

Replication and extension of Alicke (1985): Supplementary 35
Figure S7. Normal Q-Q plot for commonness predicted from desirability and other ratings
before exclusion.

Replication and extension of Alicke (1985): Supplementary 36
After Exclusion
Figure S8. Residuals versus fitted plot for self-minus-other ratings predicted from desirability
and controllability after exclusion.

Replication and extension of Alicke (1985): Supplementary 37
Figure S9. Normal Q-Q plot for self-minus-other ratings predicted from desirability and
controllability after exclusion

Replication and extension of Alicke (1985): Supplementary 38
Figure S10. Residuals versus fitted plot for commonness predicted from desirability and self-
ratings after exclusion.

Replication and extension of Alicke (1985): Supplementary 39
Figure S11. Normal Q-Q plot for commonness predicted from desirability and self-ratings
after exclusion.

Replication and extension of Alicke (1985): Supplementary 40
Figure S12. Residuals versus fitted plot for commonness predicted from desirability and other
ratings after exclusion.

Replication and extension of Alicke (1985): Supplementary 41
Figure S13. Normal Q-Q plot for commonness predicted from desirability and other ratings
after exclusion.

Replication and extension of Alicke (1985): Supplementary 42
Exploratory Analyses
The correlation comparison between desirability and self, and desirability and other was pre-
registered as one of the main analyses, whereas the remaining two correlation correlations
were beyond our pre-registration. We report the results and the effect sizes of all three
comparisons in this section, so that readers can compare the strengths of these differences.
We compared correlations between the study variables, using the R package cocor
(Diedenhofen & Musch, 2015). Since the package is limited to comparisons of only two
correlations, we focused on only the main effects for the hypotheses involving more than one
predictor. Comparisons were based on dependent groups with overlapping variables. Using
the results, effect sizes were computed using Lakens’ (2013) spreadsheet calculator. Results
are summarized in Table 19.
Table S14. Summary of correlation comparisons and effect sizes
Correlations
r.jk – r.jh
t
p
q
Desire, self |
Desire, other***
.31 [.22, .42]
10.67
<.001 (one-tailed)
0.88
Self-minus-other, desire |
Self-minus-other,
control***
.79 [.62, .96]
10.72
<.001 (one-tailed)
1.04
Common, desire |
Common, self-minus-
other***
.42 [.32, .53]
11.34
<.001
0.76
Observations: 149
Note. r.jk – r.jh refers to the difference between the correlations. r.kh refers to the related
correlation. *** indicates p <.001. Hendrickson, Stanley, and Hills' (1970) t values are
reported. Hittner, May and Silver’s (2003) z values are reported. q indicates Cohen’s q, an
effect size used for measuring correlational difference.

Replication and extension of Alicke (1985): Supplementary 43
Results after Exclusion
The below tables summarize the results after excluding data that met our pre-registered
criteria. For details of the criteria, please refer to the Replication Recipe.
The full exclusion criteria apply to all conditions, except the “Other ratings” condition in the
second-wave sample. For this condition, we removed failure to pass attention checks from the
criteria given an error in the Qualtrics survey. For the attention checks of this condition,
“very common” and “not at all common” were used instead of “very characteristic of the
average American” and “not at all characteristic of the average American”. Since this error
undermined the attention checks, participants were only excluded if they met the other
exclusion criteria, such as English proficiency and seriousness towards the survey.
Table S15. Summary of demographics of the first-wave and second-wave samples after
exclusion
First-wave
(n = 607)
Second-wave
(n = 771)
Gender
Male
309 (50.9%)
346 (44.9%)
Female
294 (48.4%)
423 (54.9%)
Missing
4 (0.7%)
2 (0.3%)
Age
Mean (SD)
39.3 (12.1)
39.5 (12.4)
Median [Min, Max]
36.0 [18.0, 77.0]
37.0 [18.0, 87.0]
Missing
4 (0.7%)
2 (0.3%)

Replication and extension of Alicke (1985): Supplementary 44
Table S16. Means, standard deviations, and correlations with confidence intervals after exclusion
Variable
M
SD
Desirability
Controllability
Commonness
Self-ratings
Other-ratings
Desirability
3.73
1.78
Controllability
4.94
0.91
.01
[-.15, .17]
Commonness
4.07
0.54
.64**
.21*
[.54, .73]
[.05, .36]
Self-ratings
3.73
1.28
.92**
.02
.61**
[.89, .94]
[-.14, .18]
[.50, .70]
Other-ratings
4.23
0.85
.02
.06
.52**
.01
[-.14, .18]
[-.10, .22]
[.39, .63]
[-.15, .17]
Self-minus-other ratings
-0.50
1.53
.76**
-.02
.22**
.83**
-.55**
[.68, .82]
[-.18, .14]
[.06, .37]
[.77, .88]
[-.65, -.42]
Note. M and SD are used to represent mean and standard deviation, respectively. Values in square brackets indicate the 95% confidence interval
for each correlation. The confidence interval is a plausible range of population correlations that could have caused the sample correlation
(Cumming, 2014). * indicates p < .05. ** indicates p < .01.

Replication and extension of Alicke (1985): Supplementary 45
Figure S14. Scatterplot showing the relationship between desirability and self-minus-other ratings with 95% confidence interval after exclusion.

Replication and extension of Alicke (1985): Supplementary 46
Table S17
Regression results using self-minus-other ratings as the criterion after exclusion
Predictor
b
b
95% CI
[LL, UL]
beta
beta
95% CI
[LL, UL]
sr2
sr2
95% CI
[LL, UL]
Fit
Difference
(Intercept)
-0.24***
[-0.36, -0.13]
R2 = .59***
Desirability
0.46***
[0.40, 0.53]
0.77
[0.66, 0.87]
.59
[.48, .69]
95% CI [.48,.66]
Controllability
-0.07
[-0.19, 0.05]
-0.06
[-0.16, 0.04]
.00
[-.01, .02]
F(2, 146) = 104.7***
(Intercept)
-0.25***
[-0.35, -0.14]
R2 = .66***
ΔR2 = .07***
Desirability
0.45***
[0.39, 0.51]
0.74
[0.64, 0.84]
.54
[.43, .65]
95% CI [.56,.71]
F(3, 145) = 91.9***
95% CI [.02, .12]
F(1, 145) = 27.88***
Controllability
-0.12*
[-0.24, -0.01]
-0.11
[-0.20, -0.01]
.01
[-.01, .03]
Interaction
0.17***
[0.10, 0.23]
0.26
[0.16, 0.36]
.07
[.02, .12]
Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. b represents unstandardized regression
weights. beta indicates the standardized regression weights. sr2 represents the semi-partial correlation squared. r represents the zero-order
correlation. LL and UL indicate the lower and upper limits of a confidence interval, respectively.
* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

Replication and extension of Alicke (1985): Supplementary 47
Table S18. Regression results using commonness as the criterion after exclusion
Predictor
b
b
95% CI
[LL, UL]
beta
beta
95% CI
[LL, UL]
sr2
sr2
95% CI
[LL, UL]
Fit
Difference
(Intercept)
4.07***
[3.99, 4.15]
R2 = .39***
Desirability
0.17**
[0.05, 0.28]
0.47
[0.14, 0.79]
.03
[-.01, .08]
95% CI [.27, .49]
Self-ratings
0.08
[-0.08, 0.24]
0.17
[-0.15, 0.50]
.00
[-.01, .02]
F(2, 146) = 47.35***
(Intercept)
4.07***
[3.93, 4.20]
R2 = .39***
ΔR2 = .000
Desirability
0.17**
[0.05, 0.28]
0.47
[0.14, 0.80]
.03
[-.01, .08]
95% CI [.26, .49]
F(3, 145) = 31.35***
95% CI [-.00, .00]
F(1, 145) = 0.002
Self-ratings
0.08
[-0.08, 0.24]
0.17
[-0.16, 0.50]
.00
[-.01, .02]
Interaction
-0.00
[-0.05, 0.05]
-0.00
[-0.13, 0.13]
.00
[-.00, .00]
Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. b represents unstandardized regression
weights. beta indicates the standardized regression weights. sr2 represents the semi-partial correlation squared. r represents the zero-order
correlation. LL and UL indicate the lower and upper limits of a confidence interval, respectively.
* indicates p < .05. ** indicates p < .01. *** indicates p < .001.

Replication and extension of Alicke (1985): Supplementary 48
Table S19. Regression results using commonness as the criterion after exclusion
Predictor
b
b
95% CI
[LL, UL]
beta
beta
95% CI
[LL, UL]
sr2
sr2
95% CI
[LL, UL]
Fit
Difference
(Intercept)
4.07***
[4.02, 4.11]
R2 = .81**
Desirability
0.04**
[0.01, 0.08]
0.13
[0.04, 0.22]
.01
[-.00, .02]
95% CI [.75,.84]
F(2, 146) = 309.7***
Other ratings
0.89***
[0.79, 0.99]
0.82
[0.73, 0.91]
.42
[.31, .53]
(Intercept)
4.03***
[3.97, 4.09]
R2 = .81**
ΔR2 = .003
Desirability
0.04*
[0.00, 0.07]
0.10
[0.01, 0.20]
.01
[-.00, .02]
95% CI [.76,.85]
F(3, 145) = 209.5***
95% CI [-.00, .01]
F(1, 145) = 2.53
Other ratings
0.92***
[0.81, 1.02]
0.84
[0.75, 0.94]
.40
[.29, .51]
Interaction
0.05
[-0.01, 0.11]
0.06
[-0.01, 0.14]
.00
[-.00, .01]
Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. b represents unstandardized regression
weights. beta indicates the standardized regression weights. sr2 represents the semi-partial correlation squared. r represents the zero-order
correlation. LL and UL indicate the lower and upper limits of a confidence interval, respectively.
* indicates p < .05. ** indicates p < .01. *** indicates p < .001