Personality Profiles of Effective Leadership Performance in Assessment Centers

Human Performance
ISSN: 0895-9285 (Print) 1532-7043 (Online) Journal homepage: http://www.tandfonline.com/loi/hhup20
Personality profiles of effective leadership

performance in assessment centers
Alissa Denzin Parr, Stephanie T. Lanza & Paul Bernthal
To cite this article: Alissa Denzin Parr, Stephanie T. Lanza & Paul Bernthal (2016) Personality
profiles of effective leadership performance in assessment centers, Human Performance, 29:2,
143-157, DOI: 10.1080/08959285.2016.1157596
To link to this article: http://dx.doi.org/10.1080/08959285.2016.1157596
Published online: 04 Apr 2016.
Submit your article to this journal
Article views: 70
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=hhup20
Download by: [University of Nebraska, Lincoln] Date: 01 June 2016, At: 10:25
HUMAN PERFORMANCE
2016, VOL. 29, NO. 2, 143–157
http://dx.doi.org/10.1080/08959285.2016.1157596
Personality profiles of effective leadership performance in

assessment centers
Alissa Denzin Parra, Stephanie T. Lanzab, and Paul Bernthalc
a
Select International; bPennsylvania State University; cDevelopment Dimensions International
ABSTRACT
Most research examining the relationship between effective leadership and
personality has focused on individual personality traits. However, profiles of
Downloaded by [University of Nebraska, Lincoln] at 10:25 01 June 2016
personality traits more fully describe individuals, and these profiles may be
important as they relate to leadership. This study used latent class analysis
to examine how personality traits combine and interact to form subpopula-
tions of leaders and how these subpopulations relate to performance
criteria. Using a sample of 2,461 executive-level leaders, six personality
profiles were identified: Unpredictable Leaders with Low Diligence (7.3%);
Conscientious, Backend Leaders (3.6%); Unpredictable Leaders (8.6%);
Creative Communicators (20.8%); Power Players (32.4%); and Protocol
Followers (27.1%). One profile performed well on all criteria in an assess-
ment center; remaining profiles exhibited strengths and weaknesses across
criteria. Implications and future directions for research are highlighted.
Moving into the 21st century, organizations are facing numerous challenges associated with the
changing nature of work. As organizations expand their boundaries globally, rely on technology for
communication, and develop knowledge workers to be adaptive in response to continually changing
work demands (Tannenbaum, 2002), it is apparent that organizations will need to adjust their
strategies to achieve a competitive advantage. In today’s dynamic work environment, it is recognized
that talented leaders can help overcome these challenges and pave the way toward achieving this
goal. Talented leaders can direct and guide employees, teams, and organizations to be successful and
to overcome these obstacles. Through their interaction with subordinates and stakeholders, leaders
can profoundly influence followers’ behaviors and thereby create a workforce primed for success
(Barling, Christie, & Hoption, 2010).
Initially guided by the trait approach of leadership, several studies sought to discover character-
istics that define talented leaders. Early studies produced inconsistent results (e.g., Bass, 1990; Mann,
1959; Stogdill, 1948), but there was renewed interest in discovering traits associated with successful
leaders after the emergence of the five-factor model of personality (Costa & McCrae, 1992). For
example, Judge and colleagues (2002) conducted a meta-analysis demonstrating that leadership
effectiveness was related to various specific factors of personality. Their findings revealed that
Extraversion and Conscientiousness are moderately and positively associated with leadership effec-
tiveness (Judge et al., 2002). In addition, DeRue and colleagues (2011) conducted a meta-analysis
that reiterated the importance of both leader traits and behaviors for leader effectiveness. They
concluded that traits and behavior explain about one third of the variance in leader effectiveness and
that Extraversion and Conscientiousness were consistent predictors of leader effectiveness (DeRue
et al., 2011).
Recent theoretical and methodological developments offer an opportunity to better understand
the link between personality and effective leadership (Antonakis, Day, & Schyns, 2012). Past studies
CONTACT Alissa D. Parr aparr@selectintl.com Select International, 5700 Corporate Drive, Pittsburgh, PA 15237.
© 2016 Taylor & Francis
144 PARR ET AL.
have typically examined personality traits individually (e.g., Bass, 1990; Mann, 1959; Stogdill, 1948),
despite the fact that traits do not exist in isolation. Alternatively, approaching personality as a higher
order combination, or cluster, of personality traits allows for a more holistic view of the individual
(Barrick & Mount, 2005). Clusters have been shown to be relevant in other areas of leadership as
well. For example, Borman and Brush (1993) created a taxonomy of leadership performance by
clustering critical performance dimensions, revealing a more complete picture of effective leadership
performance. Furthermore, past studies have primarily focused on examining how personality relates
to overall performance (see Judge et al., 2002). However, leadership performance contains many
elements that differentially relate to aspects of personality, and certain types of leaders may excel in
terms of some criteria but not in others.
The present study sought to characterize different classes of leaders based on their personality
profile and to determine how these classes relate to unique performance criteria measured in an
assessment center. Specifically, we used a novel, person-oriented approach, latent class analysis
(LCA; Collins & Lanza, 2010), to identify subgroups of leaders defined by their personality profiles.
We then conducted analyses to assess how membership in different personality classes relates to
assessment center performance dimensions, including Defining the Strategy, Executing the Strategy,
and Building Partnerships and Translating the Message. This study sought to address gaps in our
understanding of the complex link between leadership personality and performance.
Trait approach to leadership

Beginning with the “great man” hypothesis (Carlyle, 1907), several attempts have been made to
characterize an extraordinary leader. The trait approach to leadership seeks to define personality
characteristics that are related to leadership effectiveness. Personality traits differentiate individuals
based on their tendencies to think, feel, and behave (Ones, Viswesvaran, & Dilchert, 2005) and
therefore can help to elucidate why a leader may be more or less successful.
Prior studies examining personality traits have shown inconsistent and sometimes null findings
(e.g., Bass, 1990; Mann, 1959; Stogdill, 1948). This has led to a perhaps false consensus that
personality traits were poor predictors of leader outcomes (Antonakis et al., 2012). One reason for
this may be the inconsistent use of labels to identify underlying personality traits (Judge et al., 2002).
In an attempt to resolve these issues and synthesize across previous studies, Judge and colleagues
(2002) conducted a meta-analysis examining how personality relates to leadership effectiveness using
the five-factor model of personality as a framework. Overall, Extraversion emerged as the most
consistent correlate of leadership effectiveness. This is similar to other studies that demonstrated that
leaders tend to be sociable and dominant (e.g., Lord, DeVader, & Alliger, 1986; Mann, 1959; Stogdill,
1948, 1974). Furthermore, Judge and colleagues (2002) also found that Conscientiousness and
Openness to Experience were strong correlates of leadership. It follows that leadership effectiveness
may result in part from a leader being organized and receptive to hearing various perspectives before
making decisions.
In addition, DeRue and colleagues (2011) reiterated the importance of individual differences in
predicting leader effectiveness. In a meta-analysis, they found that variability in leader character-
istics, including gender, intelligence, and the Big Five, accounted for 22% of the variance in leader
effectiveness. The majority of this variance was attributed to Extraversion and Conscientiousness,
thereby highlighting the importance of these personality factors (DeRue et al., 2011).
Although the majority of research has focused on how personality traits relate to positive
outcomes, it is often the case that leaders derail on the job by not living up to their full potential.
It has been estimated that approximately 47% of leaders end up derailing (Hogan, Hogan, & Kaiser,
2010). Across several studies, personality defects have been shown to be related to leader derailment,
including being low on Extraversion, Conscientiousness, and Openness to Experience and exhibiting
a lack of integrity and honesty, egotism, and arrogance (Hogan et al., 2010; Kaiser & Hogan, 2011;
Lipman-Blumen, 2006). These studies help to provide perspective that leaders are not always
HUMAN PERFORMANCE 145
successful but that personality can be a good mechanism to determine the likelihood of success in
the role.
Although these recent meta-analyses and studies brought to the forefront the importance of
focusing on how personality relates to leadership, personality researchers have begun to stress the
need to examine not only traits individually but also personality from a more holistic perspective
(Barrick & Mount, 2005). Furthermore, several leadership researchers have called for future studies
to be more integrative and methodologically rigorous in their approaches to studying leadership
(e.g., Antonakis et al., 2012; Avolio, 2007; Bennis, 1959; DeRue et al., 2011; Hernandez, Eberly,
Avolio, & Johnson, 2011). The following section details how this study meets this call to action.
Personality profiles and leadership effectiveness: A novel approach

One method that can be used to identify personality profiles is LCA, which categorizes individuals
into latent subgroups based on multiple observed scores (Collins & Lanza, 2010). Unlike more
traditional variable-oriented approaches (e.g., multiple regression), LCA recognizes heterogeneity in

the population and accounts for that heterogeneity by identifying a set of underlying subgroups of
individuals who share common characteristics (Lanza, Collins, Lemmon, & Schafer, 2007). LCA has
been highlighted as a sophisticated and robust tool appropriate for organizational research. The
benefits of LCA include its flexibility, its ability to be used for confirmatory or exploratory purposes,
and its ability to accommodate measurement error (Wang & Hanges, 2011). Overall, LCA was noted
as a very promising method to apply in a wide range of questions in organizational research (Wang
& Hanges, 2011). In the current study, we used LCA to identify a set of personality profiles in a
sample of leaders and estimated the association between these profiles and effectiveness in an
assessment center.
Research questions
The research questions guiding the study were as follows: (a) What different personality profiles can
be identified in a sample of leaders? (b) What is the prevalence rate for each leadership profile? (c)
How do these leadership profiles relate to various performance criteria measured in an assessment
center? This study was primarily descriptive and investigative, therefore specific hypotheses were not
stated.
Method
Participants and procedures
Data used in this study were collected by an external consulting firm. Participants were 2,461
executive-level leaders from various organizations across the United States. All participants were
identified as being at least second- or third-level leaders (e.g., directors, vice presidents) and reported
an average of 14 years of experience in leadership roles. The majority of the sample was male (75%)
and White (80%). The average age was 44 (SD = 7).
In addition to this core sample, an independent sample was used to validate the latent class
structure. Participants in this sample were 5,997 leaders from organizations across different types of
industries. This sample comprised both lower level and higher level leaders. As such, this sample
included a broader array of leaders and offered an adequate test of the identified latent classes.
Participants completed a personality inventory. Subsequently, everyone participated in a day-in-
the-life assessment center. Throughout the assessment, participants assumed the role of an executive
in a fictitious organization to enact simulated work performance. They engaged in a series of visual
and written exercises, including role-plays, presentations, and in-baskets. Trained assessors rated
participants on various performance competencies after the participants performed all of the
146 PARR ET AL.
activities over the course of a full day. The trained assessors made independent ratings and then had
an integration session to finalize all the competency ratings.
Materials
Personality
The Hogan Personality Inventory (HPI; Hogan & Hogan, 2007) was used to assess personality. The
HPI contains 206 items that are scored as true or false. The HPI results in seven primary scales:
Adjustment, Ambition, Sociability, Interpersonal Sensitivity, Prudence, Inquisitiveness, and Learning
Orientation. The HPI is often used in personnel selection and places an emphasis on constructs
relevant to job, career, and occupational performance (Hogan & Hogan, 2007). Thus, this personality
inventory is appropriate to use in work settings and leadership research. Due to the proprietary
nature of the test, we were not able to calculate the internal consistency of the factors, but the HPI
manual reports alpha coefficients of .89, .86, .83, .71, .78, .78, and .75 for Adjustment, Ambition,
Sociability, Interpersonal Sensitivity, Prudence, Inquisitiveness, and Learning Orientation, respec-

tively (Hogan & Hogan, 1995).
Performance
Work performance (i.e., leadership effectiveness) was simulated by having participants engage in a
series of real-life activities that are common to leaders. Eleven performance competencies were rated
as part of the standard assessment procedure used by the external consulting company implementing
the assessment center. The performance competencies are related to dimensions that are necessary
for effective leadership, including gaining influence, coaching subordinates, making decisions,
developing business strategies, and understanding financial data. Experienced assessors rated each
competency on a scale from 1 (highly ineffective) to 3 (highly effective).
The performance competencies were subjected to a factor analysis using promax rotation. Results
suggested a three-factor solution with 45% of the variance explained. Based on the factor loadings,
the three factors were labeled Defining the Strategy, Executing the Strategy, and Building
Partnerships and Translating the Message. Defining the Strategy is composed of competencies
related to understanding financial information and business trends, as well as selecting strategies
to drive organizational growth. Executing the Strategy contains competencies related to making day-
to-day decisions, driving plans to achieve goals, initiating change, and building an environment in
which change can occur. Finally, Building Partnerships and Translating the Message is composed of
competencies related to communicating effectively, persuading others, coaching others, and net-
working. Performance factor scores were computed by taking the average of each performance
competency measured in the assessment center. All competencies were equally weighted when
calculating the overall performance factor. Reliabilities for the three factors ranged from .50 to .55.
Data preparation and analyses

The first step in data analysis involved cleaning and preparing the personality data. First, we chose to
base analyses on the five-factor model of personality because this is widely used throughout
personality and industrial/organizational psychology research. The HPI parallels the five-factor
model of personality (Digman, 1990; Hogan & Holland, 2003), with the exception that in the HPI,
Extraversion is broken down into Ambition and Sociability and Openness to Experience is broken
down into Inquisitiveness and Learning Orientation. There is both empirical and conceptual support
for creating both factors (Hogan & Hogan, 1995). Furthermore, extensive research has been con-
ducted to demonstrate the relationship between the HPI and the five-factor model of personality as
measured by various inventories (e.g., Goldberg’s Big Five factor markers, the NEO Personality
Inventory–Revised, the Interpersonal Adjective Scales; as reported by Hogan & Holland, 2003). More
recently, Salgado, Moscoso, and Alonso (2013) showed that the factor structure of the HPI resembles
the five-factor model. Thus, we felt it was appropriate to use the five-factor model terminology for
consistency. Ambition and Sociability scales were combined to form an Extraversion scale, and
Inquisitiveness and Learning Orientation scales were combined to form an Openness to Experience
scale. The reliabilities for the composite scales were .56 and .49 for Extraversion and Openness to
Experience, respectively.
Second, to prepare the data for LCA, personality scale scores were transformed into three-level
indicators. Transforming the personality scale scores into categories facilitates interpretation of the
profiles and enables a more meaningful way to examine differences between profiles. Using the HPI
user manual as a guide, scores were categorized as high (above the 65th percentile), moderate
(between the 36th and 64th percentiles), or low (below the 35th percentile). This categorization
provides more differentiation, especially when comparing those scoring high and low on the
personality variables.
Following data preparation, analyses were run in SAS 9.3. Latent class models with one through
seven classes were run using SAS PROC LCA (Lanza Dziak, Huang, Wagner, & Collins, 2013). For
each model, we used 100 random sets of starting values to ensure that the model was identified.
Selection was conducted using fit statistics (e.g., G2, Akaike information criterion [AIC], Bayesian
information criterion [BIC]), as well as interpretability of the latent classes. Furthermore, to ensure
the stability and validity of our latent class structure, a validation analysis was performed using a
separate sample of leaders. Similar procedures were used to find this solution. Finally, using the
selected latent class model for our core sample, each performance factor was examined as a distal
outcome in the latent class model to determine how personality latent class membership predicted
performance. The LCA_distal SAS macro (Lanza, Tan, & Bray, 2013) was used for this part of the
analysis.
Results
Descriptive statistics for the personality traits and performance factors are shown in Tables 1, 2, and
3, respectively. Table 1 presents means, standard deviations, and intercorrelations among all the
personality traits and performance competencies. The intercorrelations among the personality traits
are low to moderate, which is typical (Schmitt, 2014; Van Der Linden, Te Nijenhuis, & Bakker,
2010). Table 2 shows the proportion of individuals coded as low, moderate, and high on each
personality trait. A majority of the sample was identified as high on both Extraversion and
Conscientiousness, which have been shown to be important for leadership (e.g., DeRue et al.,
2011; Judge et al., 2002). Table 3 presents the means and intercorrelations among the performance
factors. On average, individuals were rated as moderately effective or less than effective (range =
1.86–2.00; see Table 2), which is typical of assessment center scores because the majority of
assessment centers are geared toward development. Furthermore, the correlations among factors
ranged from 0.11 to 0.34 (see Table 3), indicating that these are mostly distinct performance factors.
Model selection
Model fit information for the models with one through seven latent classes are shown in Table 4.
The AIC was smallest, indicating an optimal balance between fit and parsimony, for seven classes,
whereas the BIC was smallest for four classes. After careful inspection of models with four to seven
classes, we selected the six-class solution, which had low fit statistics and good model interpretability
(G2 = 160.95, AIC = 290.95, BIC = 668.49, df = 177).
Table 5 shows the prevalence of each latent class and, for each latent class, the probability of
having low, moderate, or high levels of each personality trait. Latent classes were interpreted and
labeled based on these item-response probabilities. Individuals in the first latent class were char-
acterized by low Emotional Stability, Agreeableness, and Conscientiousness (0.76, 0.83, and 0.98,
respectively), and this class was therefore labeled as Unpredictable Leaders with Low Diligence.
148
PARR ET AL.
Table 1. Means, standard deviations, and intercorrelations for personality and performance criteria.
Factor M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. Emotional Stability 63.33 25.26
2. Agreeableness 55.32 29.88 .39**
3. Conscientiousness 59.81 25.62 .46** .26**
4. Extraversion 65.58 19.75 .25** .33** −.05*
5. Openness to Experience 60.31 21.15 .14** .13** −.05* .36**
6. Strategic Thinking 1.85 .72 .02 −.03 −.05* .10** .22**
7. Financial Acumen 1.89 .74 .06 −.01 .00 .10** .17** .38**
8. Decision Making 2.26 .56 .05* −.02 .00 .07** .12** .21** .26**
9. Execution 2.16 .62 .01 .02 .04 .07** .07** .24** .23** .30**
10. Empowerment 1.71 .59 .05 .01 −.06 .04 .05 .08 .04 .36** .11**
11. Building Talent 1.80 .65 .01 .06* .04 −.01 .00 .12** .04 .10** .15** .09*
12. Leading Change 1.87 .59 −.01 .06** −.03 .07** .10** .30** .15** .22** .22** .15** .19**
13. Building Networks 2.00 .65 .04 .10** −.01 .11** .07** .12** .06* .15** .02 .11** .11** .16**
14. Influence 1.87 .66 .06 .11** .03 .20** .11** .12** .16** .21** .11** .16** .08* .16** .34**
15. Coaching 1.91 .65 .02 .08** .00 .02 .02 .08** .02 .11** .04 .10** .11** .15** .17** .18**
16. Communication 2.10 .33 .02 .06* .02 .07** .10** .18** .12* .18** .07** .09* .07 .10** .17** .22** .16**
Note. N = 2,461. *p<.05. **p<.01.
Table 2. Descriptive statistics for personality traits.

Indicators of Latent Class Level Frequency (%)
Emotional Stability Low 434 (17.6%)
Moderate 685 (27.8%)
High 1,342 (54.5%)
Agreeableness Low 655 (26.6%)
High 877 (35.6%)
Conscientiousness Low 559 (22.7%)
High 1,234 (50.1%)
Extraversion Low 205 (8.3%)
High 1,404 (57.3%)
Openness Low 346 (14.2%)
High 1,106 (45.2%)
Note. N = 2,461.
Table 3. Means, standard deviations, and intercorrelations for performance indicators.

Performance Factor M SD 1 2 3
1. Defining the Strategy 1.86 .64 .55
2. Executing the Strategy 2.00 .39 .34** .52
3. Building Partnerships & Translating the Message 1.98 .43 .11** .21** .50
Note. N = 2,461. Internal consistency values (Cronbach’s alphas) appear in bold along the diagonal.
**p < .01.
Table 4. Comparison of baseline models.

No. of Classes df Likelihood Ratio G2 AIC BIC Entropy R2 Solution %
1 232 1467.43 1487.43 1545.51 1.00 100
2 221 666.81 708.81 830.79 0.58 100
3 210 347.83 411.83 597.70 0.66 100
4 199 244.43 330.43 580.19 0.63 100
5 188 187.94 295.94 609.59 0.63 94
6 177 160.95 290.95 668.49 0.67 19
7 166 138.83 290.83 732.27 0.65 18
Note. N = 2,461. Boldface type indicates the selected model. Solution % is the percentage of times solution was selected out of 100
random sets of starting values. AIC = Akaike’s information criterion; BIC = Bayesian information criterion.
Approximately 7.3% of individuals belonged to this class. Individuals in the second class, labeled
Conscientious, Backend Leaders (3.6%), were likely to be high on Conscientiousness (0.95) but low
on Agreeableness (0.66) and Extraversion (0.86). The third latent class (8.6%) was labeled
Unpredictable Leaders because they had high probabilities of low Emotional Stability and
Agreeableness (0.51 and 0.61, respectively). Individuals in the fourth latent class (20.8%), labeled
Creative Communicators, had high probabilities of high Extraversion (0.74) and Openness to
Experience (0.60) but low Conscientiousness (0.56). The fifth latent class (32.4%) was labeled
Power Players because members of this class had high probabilities for high scores on all five
personality traits. This was the most common personality profile among the leaders. Finally, the sixth
latent class (27.1%), Protocol Followers, comprised individuals who were likely to be high on
Emotional Stability and Conscientiousness (0.71 and 0.74, respectively).
150 PARR ET AL.
Table 5. Item-Response probabilities for six-class model: probability of endorsing item given latent class.
Latent Classes
1 (7.3%) 2 (3.6%) 3 (8.6%) 4 (20.8%) 5 (32.4%) 6 (27.1%)
Low
Emotional Stability .76 .18 .51 .26 .03 .02
Agreeableness .83 .66 .61 .27 .02 .24
Conscientiousness .98 .04 .00 .56 .09 .03
Extraversion .20 .86 .36 .00 .01 .04
Openness .22 .40 .28 .08 .04 .20
Moderate
Agreeableness .05 .16 .34 .54 .38 .38
Extraversion .51 .05 .47 .26 .01 .77
Openness .43 .52 .49 .32 .33 .51
High
Agreeableness .12 .19 .05 .19 .60 .38

Extraversion .29 .09 .17 .74 .98 .21
Openness .34 .07 .23 .60 .63 .29
Note. N = 2,461. Probabilities greater than .50 in bold to facilitate interpretation.
1 = Unpredictable Leader with Low Diligence; 2 = Conscientious, Backend Leader; 3 = Unpredictable Leader; 4 = Creative
Communicator; 5 = Power Player; 6 = Protocol Follower.
Table 6. Validation results: item-response probabilities and class membership probabilities for six-class model.
Latent Classes
1 (7.5%) 2 (9.1%) 3 (16.8%) 4 (27.7%) 5 (25.3%) 6 (13.6%)
Low
Agreeableness .88 .85 .55 .36 .06 .30
Extraversion .29 .99 .00 .01 .00 .17
Openness .25 .41 .19 .09 .07 .25
Moderate
Agreeableness .11 .12 .36 .49 .41 .43
Extraversion .43 .01 .99 .11 .14 .75
Openness .48 .47 .47 .41 .34 .55
High
Agreeableness .01 .03 .09 .15 .52 .28
Extraversion .28 .00 .01 .88 .86 .07
Openness .27 .13 .34 .50 .59 .20
Note. N = 5,997. Probabilities greater than .50 in bold to facilitate interpretation.
To validate the six-class solution, we replicated this analytic procedure using an independent
sample of leaders. LCA results using the validation sample also suggested a six-class solution
(G2 = 238.50, AIC = 368.50, BIC = 803.94, df = 177). Despite the fact that class prevalences were
slightly different across samples, the characteristics of the classes remained essentially the same (see
Table 6). Thus, we felt confident that our six-class solution was valid and could generalize to a
broader population of leaders.
Class membership and performance

To test whether and how class membership predicts performance, each performance factor was
separately included in the latent class model as a distal outcome. Omnibus tests revealed that class
membership was significantly associated with all three performance factors (p < .001 for each).
Table 7 shows the standardized mean differences of performance outcomes conditional on person-
ality latent class membership. To compute the standardized mean differences, we subtracted the
lowest average score on each performance factor to the other scores within that factor. For the
performance factors Defining the Strategy and Executing the Strategy, Conscientious, Backend
Leaders, and Power Players had the highest means, and Protocol Followers and Unpredictable
Leaders had the lowest means. Thus, Conscientious, Backend Leaders and Power Players showed
high performance on defining and implementing strategies. For the performance factor, Building
Partnerships and Translating the Message, Creative Communicators, and Power Players had highest
means, whereas Conscientious, Backend Leaders and Unpredictable Leaders with Low Diligence had
the lowest means. As such, Creative Communicators and Power Players demonstrated relatively high
performance in terms of communicating with others and building networks.

We then examined the data from a different perspective by looking at performance within each
personality profile to elucidate strengths and weaknesses of each type of leader. For example, Power
Players scored relatively high on all performance factors. In comparison to their peers, they had
either the highest or second highest performance score on each factor. Alternatively, Unpredictable
Leaders performed more poorly in Defining the Strategy and Executing the Strategy and moderately
in Building Partnerships and Translating the Message, relative to their peers. Other classes tend to
have more complex relationships. For example, Conscientious, Backend Leaders exceled in Defining
the Strategy and Executing the Strategy but showed the worst performance in Building Partnerships
and Translating the Message. In addition, Creative Communicators performed well on Building
Partnerships and Translating the Message but scored only moderately on the other performance
factors.
Discussion
The purpose of this study was to characterize different profiles of leaders based on their composite
personality structure using a novel, person-oriented statistical approach and to determine how
individuals with these leader profiles perform on competencies important for leadership as measured
in an assessment center. The results of the current study offer some unique contributions in a few
areas. First, results demonstrated that there are six different profiles of leaders when examining their
personality composite. Although previous research has examined individual personality factors of
leaders, this study is one of the first to examine personality holistically. Furthermore, this exemplifies
that there is not a “one-size-fits-all” personality model for leadership; rather, there are several
different subpopulations of leaders based on their composite personality structure.
Table 7. Results showing performance outcomes conditional on personality latent class membership.
Latent Classes
Unpredictable Leader Conscientious, Unpredictable Creative Power Protocol
Performance Factor with Low Diligence Backend Leader Leader Communicator Player Follower
Defining the Strategy 0.25 0.53 0.08 0.37 0.40 0.00
Executing the Strategy 0.21 0.46 0.00 0.22 0.29 0.15
Building Partnerships and 0.13 0.00 0.19 0.27 0.29 0.22
Translating the Message
Note. N = 2,461. Standardized means were computed based on subtracting the lowest mean score from the other scores within
each performance factor.
152 PARR ET AL.
Another contribution is that our results showed how these different profiles of leaders perform on
competencies necessary for leadership. Much can be gained by examining these classes and their
performance. The most prevalent class, Power Players, on average performed well on all perfor-
mance factors. Members of this class also tend to be emotionally stable, agreeable, conscientious,
social, and open to new ideas and therefore are likely to appeal to a broad audience. Given their
charismatic nature and business savvy skills, they may excel in a wide range of situations. Because
Power Players tended to be high on all the personality dimensions, there may be some concern as to
whether participants were responding in a socially desirable manner. However, research has demon-
strated that social desirability is not a great concern in real-life work settings and that relationships
between personality and job performance do not appear to be suppressed by social desirability
(Hough, 1998; Hough, Eaton, Dunnette, Kamp, & McCloy, 1990; Hough & Oswald, 2008; Ones,
Viswesvaran, & Reiss, 1996; Oswald & Hough, 2011). Furthermore, it is possible that these indivi-
duals may be better at impression management, which is important for effective leadership.
Protocol Followers, the second most prevalent class, tended to be emotionally stable and con-
scientious yet performed lower on Defining the Strategy and Executing the Strategy. This class of
leaders, on average, was moderately good at forging interpersonal relationships and developing
others, but they do not have the skills needed to identify future directions for the organization
and drive those changes.
The third most prevalent class, Creative Communicators, tended to be very social and open to
new ideas but somewhat less diligent in their actions. Their personality tendencies come across in
their performance as they exceled in Building Partnerships and Translating the Message but were
less skilled in developing and implementing strategies. Thus, this class of leaders may be instru-
mental in gaining appeal from followers for new innovations but because they are less diligent, they
may not do as well in strategy formation and implementation.
The final three classes demonstrated more complex relationships with the performance factors.
Unpredictable Leaders, who tended to be less stable and agreeable, did not perform as well in
formulating and implementing strategies. In addition, Conscientious, Backend Leaders, who tended
to be stable and diligent yet less agreeable and social, were adept at developing and acting on
strategies but did not excel at conveying the message or gaining mass appeal. These leaders are much
more pragmatic and less likely to come across as charismatic and persuasive. Finally, Unpredictable
Leaders with Low Diligence lacked skills in networking and communication and also tended to be
less emotionally stable, agreeable, and conscientious. In turn, they may not be perceived as appealing
and therefore may be unable to win relationships and persuade others.
Despite the sample being executive leaders and having several years of experience in management
positions, the six profiles varied in the degree to which they were effective in the three performance
factors. The magnitude between the performance factors varied relative to each other suggesting
leaders can be effective in one area but not necessarily another area. All of these leaders have learned
how to be at least minimally effective in their roles in order to advance in their careers. Therefore, it
is quite possible that these leaders may have learned how to successfully leverage their strengths and
downplay their weaknesses. In addition, leadership is not only about the individual. Rather, leaders
are a part of a broader team and organization. Leaders who are able to surround themselves by
individuals who can make up for their limitations may be more successful on the job.
Implications
The results from this study have several theoretical and practical implications for leadership and
personality research. Regarding theoretical implications, this study helped to elucidate how person-
ality, when examined holistically, relates to assessment center-based leadership. There have been
several calls to examine combinations of personality traits, as opposed to studying traits individually
(Barrick & Mount, 2005); this study answers this call. Furthermore, this research found that there is
not one profile of a leader. Rather, many different personality compositions make up leaders, some
of which are more effective than others. Each profile has different strengths and weaknesses across a
range of performance criteria. Examining personality holistically, and how personality profiles relate
to multiple performance criteria, helped to tease apart this complex relationship.
Regarding practical implications, this study highlights some of the important personality
patterns of leaders and their resulting performance. These combinations can then be used as a
supplement for personnel selection. Although the use of personality for selection has been met
with some controversy, if personality is matched for the purposes of the specific situation, then it
may add value to the selection system (e.g., Morgeson et al., 2007a, 2007b; Murphy &
Dzieweczynski, 2005; Ones, Dilchert, Viswesvaran, & Judge, 2007). For example, if an organiza-
tion is seeking a leader who can build networks and gain acceptance from employees on new
changes occurring, then it may be important to focus on finding Creative Communicators.
Alternatively, if an organization is seeking a leader who can develop a new vision for the future
and execute that vision, then an organization may be better suited to identify Conscientious,
Backend Leaders.
Some organizations might be tempted to use personality profiles to screen out leaders who
do not seem to fit the ideal personality profile for a specific position. Although there is some
merit to this approach, a more balanced approach combines data from measures of personality,
behavior, experience, and motivation (Barrick & Mount, 2005). This holistic view recognizes
that high levels of motivation or practice can overcome natural tendencies inherent in
personality styles. A good example would be a Conscientious, Backend leader who is intro-
verted but pushes to network with others in an effort to further one’s career. Although many
organizations seek streamlined and low-cost selection methods, they will likely miss out on
strong candidates if they rely too much on personality measures to drive their decisions.
This study also questions the conventional belief that one leader is best suited to perform all
job duties. Although one leadership class, Power Players, excelled in all performance factors,
the remainder of the classes displayed both strengths and weaknesses across the criteria. This
highlights the notion that one leader may not always be appropriate. Rather, dual leadership
may be more instrumental for the success of the organization. Although the importance of
dual leadership has been acknowledged when leading for innovation (Hunter, Cushenbery,
Fairchild, & Boatman, 2012), it may also be relevant in other situations. For example, if an
organization is unable to find a well-qualified Power Player, then it may be an equally good
alternative to hire a Conscientious, Backend Leader and a Creative Communicator.
Alternatively, a leader with specific tendencies may create a team of strong players who are
able to supplement or compensate for the leader’s shortcomings (Humphrey, Hollenbeck,
Meyer, & Ilgen, 2007). Imagine an Unpredictable Leader with Low Diligence who surrounds
himself with emotionally stable, conscientious team members. By openly sharing personality
profiles, team members can work together to counterbalance each other and stay on track. This
“Great Team” model of leadership provides an interesting contrast to those who may subscribe
to the “Great Man” hypothesis.
In addition, this study provides insights for leadership development. Knowing to what personality
class leaders belong may aid in identifying developmental opportunities for improving the leader’s
skills. For example, Conscientious, Backend Leaders should engage in developmental opportunities
that will help them to develop their communication, coaching, and persuasion skills. An awareness
of potential weaknesses in leadership performance provides an opportunity to preemptively build
and develop relevant skills.
Finally, the personality profiles in the study might help to identify leaders with high
potential who have not yet had the opportunity to demonstrate their skills. For example,
organizations can attempt to identify associates in a group of engineers with a Power Player
personality profile. This undiscovered high-potential leader could be presented with special
development opportunities that would give him or her the chance to gain experience and hone
154 PARR ET AL.
leadership skills over time. Personality profiling can help source succession management
programs by finding future leaders with the right “DNA” to fill open positions.
Study limitations
Despite these unique contributions, a few limitations should be noted. First, assessment center
ratings of performance, rather than on-the-job performance, were utilized. However, several meta-
analyses have demonstrated support for assessment centers as predictors of on-the-job performance
(e.g., Arthur, Day, McNelly, & Edens, 2003; Gaugler, Rosenthal, Thornton, & Bentson, 1987). Also,
the activities in the assessment center were created to mimic on-the-job exercises and therefore
provide a realistic view on performance. Having participants complete several exercises over the
course of a full day helped to gather multiple pieces of information about the participants’
performance.
Similarly, some of the average performance scores for the leader profiles were lower, which may
suggest to some that not all the leaders in the sample were effective. However, it is often the case that
in assessment centers focused on development, scores tend to be lower (Thornton & Rupp, 2006).
Furthermore, assessment center ratings tend to balance criterion-referenced ratings with norm-
referenced behavior. This implies that not only is performance in the assessment center being
compared to an average leader but the ratings also consider what is the “gold standard” for leader-
ship performance. For example, research shows that most leaders struggle with common skills like
execution, strategy definition, and coaching (Paese, 2013). As such, lower scores, especially in these
areas, do not necessarily indicate that the leader is less than effective but rather confirm that most
leaders struggle with some of these competencies.
In addition, the reliabilities for the performance factors were lower than optimal. Reliability,
however, generally acts as a suppressor of validity, suggesting that the results may serve as a lower
bound estimate of effects (Greer, Dunlap, & Hunter, 2006). Thus, although our results are on the
conservative side, they can be considered meaningful because lower reliabilities make it more
challenging to identify significant results.
Finally, this sample included executive-level leaders who had on average 14 years of experience.
Even though we found support that our personality profiles hold true across a broader sample of
leaders with less experience, the relationship to performance is focused solely on leader effectiveness
rather than leader emergence. It would be important to examine another sample of leaders with less
tenure to identify how emergent leaders perform on these criteria.
Future recommendations and conclusions

The results and limitations from this study highlight the need to continue research in this area.
Whereas the majority of research examining personality and leadership effectiveness has
focused on traits individually, there has been increasing attention and interest in examining
personality holistically. Leadership research would benefit from broader use of a person-
oriented approach to examine personality typologies in order to better understand leadership
profiles and their associated effectiveness. This method accepts the notion that personality is
complex and therefore may illuminate how personality traits interact with each other to form a
profile. Thus, future research using LCA to examine personality and leadership effectiveness is
warranted.
First, it would be important to replicate these results using another sample. In particular, we
encourage examining objective performance or on-the-job performance measures to see whether any
differences emerge. In addition, it would be interesting to determine whether these profiles of
leadership quantitatively or qualitatively differ across leader levels. It may be the case that executives
comprise a greater proportion of leaders with a particular personality profile compared to opera-
tional leaders. Furthermore, it may be interesting to identify whether the extent to which individual
characteristics are associated with personality profiles or, perhaps more interesting, whether links
between personality profile and leadership performance differ across gender or leader levels.
Furthermore, recognizing the complexity of leadership, it will be also important to examine how
various situational factors influence the relationship between the personality profiles and perfor-
mance. For example, climate stability or culture may impact which type of leader excels. In addition,
taking a more macroperspective, it would be interesting to learn how having different leader profiles
affects organizational performance.
In conclusion, this study provides preliminary evidence for summarizing the synergistic relation-
ship between distinct personality profiles of leaders and their associated performance on three
factors. It is essential that we recognize the many pathways to successful leadership, whether we
are considering personality or behavioral style. It is our hope that this study is a step in this direction
and, more important, that it compels others to take similar steps.
Acknowledgment
This research was supported by awards P50 DA010075 and P50 DA039838 from the National
Institute on Drug Abuse (NIDA). The content is solely the responsibility of the authors and does
not necessarily represent the official views of NIDA or the National Institues of Health.
References
Antonakis, J., Day, D. V., & Schyns, B. (2012). Leadership and individual differences: At the cusp of a renaissance. The
Leadership Quarterly, 23, 643–650. doi:10.1016/j.leaqua.2012.05.002
Arthur, W., Day, D., McNelly, T. L., & Edens, P. S. (2003). A meta-analysis of the criterion-related validity of
assessment center dimensions. Personnel Psychology, 56, 125–513. doi:10.1111/j.1744-6570.2003.tb00146.x
Avolio, B. J. (2007). Promoting more integrative strategies for leadership theory-building. American Psychologist, 62,
25–33. doi:10.1037/0003-066X.62.1.25
Barling, J., Christie, A., & Hoption, C. (2010). Leadership. In S. Zedeck (Ed.), APA handbook of industrial and
organizational psychology (Vol. 1, pp. 183–240). Washington, DC: American Psychological Association.
Barrick, M. R., & Mount, M. K. (2005). Yes, personality matters: Moving on to more important matters. Human
Performance, 18(4), 359–372. doi:10.1207/s15327043hup1804_3
Bass, B. M. (1990). Bass and Stogdill’s handbook of leadership. New York, NY: Free Press.
Bennis, W. G. (1959). Leadership theory and administrative behavior: The problem of authority. Boston, MA: Boston
University Human Relations Center.
Borman, W. C., & Brush, D. H. (1993). More progress toward a taxonomy of managerial performance requirements.
Human Performance, 6, 1–21.
Carlyle, T. (1907). On heroes, hero-worship, and the heroic in history. Boston, MA: Houghton Mifflin.
Collins, L. M., & Lanza, S. T. (2010). Latent class and latent transition analysis for applications in the social, behavioral,
and health sciences. Hoboken, NJ: Wiley & Sons.
Costa, P. T., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory
(NEO-FFI) manual. Odessa, FL: Psychological Assessment Resources.
DeRue, D. S., Hahrgang, J. D., Wellman, N., & Humphrey, S. E. (2011). Trait and behavioral theories of leadership: An
integration and meta-analytic test of their relative validity. Personnel Psychology, 64, 7–52. doi:10.1111/j.1744-
6570.2010.01201.x
Digman, J. M. (1990). Personality structure: Emergence of the five- factor model. Annual Review of Psychology, 41,
417–440. doi:10.1146/annurev.ps.41.020190.002221
Gaugler, B. B., Rosenthal, D. B., Thornton, G. C., & Bentson, C. (1987). Meta-analysis of assessment center validity.
Journal of Applied Psychology, 72, 493–511. doi:10.1037/0021-9010.72.3.493
Greer, T., Dunlap, W. P., & Hunter, S. T. (2006). The effects of skew on internal consistency. Journal of Applied
Psychology, 91, 1351–1358. doi:10.1037/0021-9010.91.6.1351
Hernandez, M., Eberly, M. B., Avolio, B. J., & Johnson, M. D. (2011). The loci and mechanisms of leadership:
Exploring a more comprehensive view of leadership theory. The Leadership Quarterly, 22, 1165–1185. doi:10.1016/j.
leaqua.2011.09.009
Hogan, J., Hogan, R., & Kaiser, R. B. (2010). Management derailment: Personality assessment and mitigation. In S.
Zedeck (Ed.), American Psychological Association handbook of industrial and organizational psychology(Vol. 3, pp.
555–575). Washington, DC: American Psychological Association.
156 PARR ET AL.
Hogan, J., & Holland, B. (2003). Using theory to evaluate personality and job-performance: A socioanalytic perspec-
tive. Journal of Applied Psychology, 88, 100–112. doi:10.1037/0021-9010.88.1.100
Hogan, R., & Hogan, J. (1995). Hogan personality inventory manual (2nd ed.). Tulsa, OK: Hogan Assessment Systems.
Hogan, R., & Hogan, J. (2007). Hogan personality inventory manual (3rd ed.). Tulsa, OK: Hogan Assessment Systems.
Hough, L. M. (1998). Effects of intentional distortion in personality measurement and evaluation of suggested
palliatives. Human Performance, 11, 209–214. doi:10.1080/08959285.1998.9668032
Hough, L. M., Eaton, N. L., Dunnette, M. D., Kamp, J. D., & McCloy, R. A. (1990). Criterion-related validities of
personality constructs and the effect of response distortion on those validities [Monograph]. Journal of Applied
Psychology, 75, 581–595. doi:10.1037/0021-9010.75.5.581
Hough, L. M., & Oswald, F. L. (2008). Personality testing and industrial-organizational psychology: Reflections,
progress, and prospects. Industrial and Organizational Psychology, 1, 272–290. doi:10.1111/iops.2008.1.issue-3
Humphrey, S. E., Hollenbeck, J. R., Meyer, C. J., & Ilgen, D. R. (2007). Trait configurations in self-managed teams: A
conceptual examination of the use of seeding for maximizing and minimizing trait variance in teams. Journal of
Applied Psychology, 92(3), 885–892. doi:10.1037/0021-9010.92.3.885
Hunter, S. T., Cushenbery, L., Fairchild, J., & Boatman, J. (2012). Partnerships in leading for innovation: A dyadic
model of collective leadership. Industrial and Organizational Psychology: Perspectives on Science and Practice, 5,
424–428. doi:10.1111/iops.2012.5.issue-4
Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and leadership: A qualitative and quantitative
review. Journal of Applied Psychology, 87, 765–780. doi:10.1037/0021-9010.87.4.765
Kaiser, R. B., & Hogan, J. (2011). Personality, leader behavior, and overdoing it. Consulting Psychology Journal: Practice
and Research, 63, 219–242. doi:10.1037/a0026795
Lanza, S. T., Collins, L. M., Lemmon, D. R., & Schafer, J. L. (2007). PROC LCA: A SAS procedure for latent class
analysis. Structural Equation Modeling: A Multidisciplinary Journal, 14(4), 671–694. doi:10.1080/
10705510701575602
Lanza, S. T., Dziak, J. J., Huang, L., Wagner, A. T., & Collins, L. M. (2013). Proc LCA & Proc LTA users’ guide (Version
1.3.0). University Park, PA: The Methodology Center, Penn State. Retrieved from http://methodology.psu.edu
Lanza, S. T., Tan, X., & Bray, B. C. (2013). Latent class analysis with distal outcomes: A flexible model-based approach.
Structural Equation Modeling: A Multidisciplinary Journal, 20, 1–26. NIHMSID: NIHMS591510. doi:10.1080/
10705511.2013.742377
Lipman-Blumen, J. (2006). The allure of toxic leaders: Why we follow destructive bosses and corrupt politicians–and how
we can survive them. Oxford, UK: Oxford University Press.
Lord, R. G., De Vader, C. L., & Alliger, G. M. (1986). A meta-analysis of the relation between personality traits and
leadership perceptions: An application of validity generalization procedures. Journal of Applied Psychology, 71, 402–
410. doi:10.1037/0021-9010.71.3.402
Mann, R. D. (1959). A review of the relationships between personality and performance in small groups. Psychological
Bulletin, 56, 241–270. doi:10.1037/h0044587
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K., & Schmitt, N. (2007a). Are we getting
fooled again? Coming to terms with limitations in the use of personality tests for personnel selection. Personnel
Psychology, 60, 1029–1049. doi:10.1111/peps.2007.60.issue-4
Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R., Murphy, K. R., & Schmitt, N. (2007b).
Reconsidering the use of personality tests in personnel selection contexts. Personnel Psychology, 60, 683–729.
doi:10.1111/peps.2007.60.issue-3
Murphy, K. R., & Dzieweczynski, J. L. (2005). Why don’t measures of broad dimensions of personality perform better
as predictors of job performance? Human Performance, 18, 343–357. doi:10.1207/s15327043hup1804_2
Ones, D., Dilchert, C., Viswesvaran, C., & Judge, T. (2007). In support of personality assessment in organizational
settings. Personnel Psychology, 60, 995–1027. doi:10.1111/j.1744-6570.2007.00099.x
Ones, D. S., Viswesvaran, C., & Dilchert, S. (2005). Personality at work: Raising awareness and correcting misconcep-
tions. Human Performance, 18(4), 389–404. doi:10.1207/s15327043hup1804_5
Ones, D. S., Viswesvaran, C., & Reiss, A. D. (1996). Role of social desirability in personality testing for personnel
selection: The red herring. Journal of Applied Psychology, 81, 660–679. doi:10.1037/0021-9010.81.6.660
Oswald, F. L., & Hough, L. M. (2011). Personality and its assessment in organizations: Theoretical and empirical
developments. In S. Zedeck (Ed.), APA handbook of industrial and organizational psychology (Vol. 2, pp. 153–184).
Washington, DC: American Psychological Association.
Paese, M. (2013). Leadership insights: A 10-year culmination of executive analytics (White Paper). Retrieved from
Development Dimensions International on June 19, 2013.
Salgado, J. F., Moscoso, S., & Alonso, P. (2013). Subdimensional structure of the Hogan personality inventory.
International Journal of Selection and Assessment, 21, 277–285. doi:10.1111/ijsa.2013.21.issue-3
Schmitt, N. (2014). Personality and cognitive ability as predictors of effective performance at work. Annual Review of
Organizational Psychology & Organizational Behavior, 1, 45–65. doi:10.1146/annurev-orgpsych-031413-091255
Stogdill, R. M. (1948). Personal factors associated with leadership: A survey of the literature. The Journal of Psychology,
25, 35–71. doi:10.1080/00223980.1948.9917362
Stogdill, R. M. (1974). Handbook of leadership. New York, NY: Free Press.

Tannenbaum, S. (2002). A strategic view of organizational training and learning. In K. Kraiger (Ed.), Creating,
implementing, and managing effective training and development (pp. 10–52). San Francisco, CA: Jossey-Bass.
Thornton, G. C., & Rupp, D. E. (2006). Assessment centers in human resource management. New York, NY: Psychology
Press.
Van Der Linden, D., Te Nijenhuis, J. T., & Bakker, A. B. (2010). The general factor of personality: A meta-analysis of
Big Five intercorrelations and a criterion-related validity study. Journal of Research in Personality, 44, 315–327.
doi:10.1016/j.jrp.2010.03.003
Wang, M., & Hanges, P. J. (2011). Latent class procedures: Applications to organizational research. Organizational
Research Methods, 14, 24–33. doi:10.1177/1094428110383988

Personality Profiles of Effective Leadership Performance in Assessment Centers

Uploaded by

Copyright:

Available Formats

Personality Profiles of Effective Leadership Performance in Assessment Centers

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Personality Profiles of Effective Leadership Performance in Assessment Centers

Uploaded by

Copyright:

Available Formats

Human Performance

ISSN: 0895-9285 (Print) 1532-7043 (Online) Journal homepage: http://www.tandfonline.com/loi/hhup20

Personality profiles of effective leadership

Alissa Denzin Parr, Stephanie T. Lanza & Paul Bernthal

To link to this article: http://dx.doi.org/10.1080/08959285.2016.1157596

Published online: 04 Apr 2016.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

Personality profiles of effective leadership performance in

Trait approach to leadership

Personality profiles and leadership effectiveness: A novel approach

traditional variable-oriented approaches (e.g., multiple regression), LCA recognizes heterogeneity in

Sociability, Interpersonal Sensitivity, Prudence, Inquisitiveness, and Learning Orientation, respec-

Data preparation and analyses

Table 2. Descriptive statistics for personality traits.

Table 3. Means, standard deviations, and intercorrelations for performance indicators.

Table 4. Comparison of baseline models.

Agreeableness .12 .19 .05 .19 .60 .38

Class membership and performance

performance in terms of communicating with others and building networks.

Future recommendations and conclusions

Stogdill, R. M. (1974). Handbook of leadership. New York, NY: Free Press.

You might also like