Deep-Level Diversity and Leadership

Abstract

In the special issue on Diversity and Leadership (April 2010), the authors made a strong case for the importance of diversity in workplace leadership, rejected premature declarations that workplace discrimination is obsolete, and called for leadership theories that acknowledge and promote the value of diversity. We appreciate all authors' stressing that the glass ceiling still exists, not only for women but for other historically low-power groups as well. We also agree that modern theories of leadership can benefit immensely from increased participation by scholars and practitioners who are not Western, White, upper-class men (Chin, 2010).

Full text

to believe that group selection can explain the
evolutionary puzzles of homosexual orienta-
tion and suicide, as claimed by Tate and
Ledbetter.
In contrast to the Winegard et al. (2010)
commentary, which proposed a plausible par-
tial explanation for male homosexual orien-
tation anchored in sexually antagonistic se-
lection, the Tate and Ledbetter (2010)
commentary proposed to explain “heterosex-
uality, homosexuality, and bisexuality simul-
taneously” (p. 930) by invoking the claim
that human genes “code for a conspecific
attraction to all members of the species” (p.
930), coupled with the erroneous assertion
that homosexuality is not “selected against”
(p. 930) “at the population level” (p. 930). It
is empirically known that the direct reproduc-
tive success of male homosexuals is substan-
tially lower than that of male heterosexuals
(Iemmola & Camperio Ciani, 2009). So un-
less there exists a reproductive benefit con-
sistently associated with genes for homosex-
uality, which may be the case, then it
certainly is selected against. That is why so
many scientists worldwide view it as an
“evolutionary puzzle.” Tate and Ledbetter,
unlike Winegard et al., provided neither a
hypothesis nor empirical evidence for a com-
pensatory benefit.
Instead, they invoked a panhuman sex-
ual attraction that is supposed to somehow
explain all forms of sexual orientation. This
claim flies in the face of all scientific evi-
dence about human sexuality. First, it does
not explain why heterosexual orientation is
highly canalized in humans, characteristic of
more than 95% of people. Second, it is con-
tradicted by the fact that most people are
attracted to sexually mature members of the
opposite sex, not in a panspecies manner to
infants, children, or those with observable
cues to low reproductive viability. Third,
most humans are not sexually attracted to
close genetic relatives, reflecting an incest
avoidance adaptation. Indeed, most people,
far from being attracted, experience disgust
and repulsion at the thought of having sex
with a close genetic relative. These are
merely three key facts from a larger body of
evidence that contradicts the notion of a pan-
human sexual attraction (Puts, 2009). In
short, Tate and Ledbetter’s (2010) claim that
“homosexuality is easily explained” (p. 930)
is amply contradicted by the existing scien-
tific evidence.
The Future of Evolutionary
Psychology
Evolutionary psychology currently provides
the most powerful current metatheory for
psychological science. It is not a static meta-
theory, but one that continues to mature and
deepen with new theoretical and empirical
advances. Progress in psychological science
requires a deep understanding of the frame-
work of evolutionary psychology and the
avoidance of common and stubbornly held
misunderstandings about it. Our original ar-
ticle was designed to clear up some of the
more common misunderstandings—an ad-
mittedly difficult task, given that (a) most
psychologists receive no formal training in
evolutionary biology, (b) evolutionary psy-
chology is widely mischaracterized in many
psychology textbooks, and (c) emotional, re-
ligious, and ideological antipathies to evolu-
tionary psychology interfere with clear, dis-
passionate evaluation. The Winegard et al.
(2010) commentary applauded us for clarify-
ing these misunderstandings. The Tate and
Ledbetter (2010) commentary badly mischar-
acterized our arguments and blithely declared
that complex psychological mysteries are
“easily explained” by invoking an outmoded
form of group selection and proposing ideas
such as a panhuman sexual attraction that are
known to be scientifically false. Their com-
mentary demonstrates that developing a deep
understanding of evolutionary psychological
science can be a formidable scholarly chal-
lenge.
REFERENCES
Alvergne, A., Faurie, C., & Raymond, M.
(2008). Developmental plasticity of human
reproductive development: Effects of early
family environment in modern-day France.
Physiology and Behavior, 95, 625– 632. doi:
10.1016/j.physbeh.2008.09.005
Buss, D. M. (2011). Evolutionary psychology:
The new science of the mind (4th ed.). Boston,
MA: Allyn & Bacon.
Buss, D. M., & Duntley, J. D. (2008). Adapta-
tions for exploitation. Group Dynamics, 12,
53– 62. doi:10.1037/1089-2699.12.1.53
Buss, D. M., & Hawley, P. (2011). The evolution
of personality and individual differences. New
York, NY: Oxford University Press.
Confer, J. C., Easton, J. E., Fleischman, D. S.,
Goetz, C., Lewis, D. M., Perilloux, C., &
Buss, D. M. (2010). Evolutionary psychology:
Controversies, questions, prospects, and limi-
tations. American Psychologist, 65, 110 –126.
doi:10.1037/a0018413
Gangestad, W. W., & Thornhill, R. (2008). Hu-
man oestrus. Proceedings of the Royal Society
of London, B, 275, 991–1000. doi:10.1098/
rspb.2007.1425
Haselton, M. G., & Buss, D. M. (2000). Error
management theory: A new perspective on
biases in cross-sex mind reading. Journal of
Personality and Social Psychology, 78, 81–
91. doi:10.1037/0022-3514.78.1.81
Haselton, M. G., & Nettle, D. (2006). The para-
noid optimist: An integrative evolutionary
model of cognitive biases. Personality and
Social Psychology Review, 10, 47– 66. doi:
10.1207/s15327957pspr1001_3
Iemmola, F., & Camperio Ciani, A. (2009). New
evidence of genetic factors influencing sexual
orientation in men: Female fecundity increase in
the maternal line. Archives of Sexual Behavior,
38, 393–399. doi:10.1007/s10508-008-9381-6
Perilloux, C., Fleischman, D. S., & Buss, D. M.
(2008). The daughter-guarding hypothesis:
Parental influence on children’s mating be-
havior. Evolutionary Psychology, 6, 217–233.
Puts, D. A. (2009). The evolution of human
sexuality: An anthropological perspective
(2nd ed.). Dubuque, IA: Kendall/Hunt.
Schmitt, D. P., & International Sexuality De-
scription Project. (2003). Universal sex differ-
ences in the desire for sexual variety: Tests
from 52 nations, 6 continents, and 13 islands.
Journal of Personality and Social Psychology,
85, 85–104. doi:10.1037/0022-3514.85.1.85
Tate, C., & Ledbetter, J. N. (2010). Oversimplify-
ing evolutionary psychology leads to explana-
tory gaps. American Psychologist,65, 929 –930.
Williams, G. C. (1966). Adaptation and natural
selection. Princeton, NJ: Princeton University.
Winegard, B., Bailey, D.H., Oxford, J., & Geary,
D. C. (2010). Trade-offs and individual dif-
ferences in evolved traits. American Psychol-
ogist,65, 928 –929.
Correspondence concerning this comment
should be addressed to David M. Buss, Depart-
ment of Psychology, University of Texas, Aus-
tin, TX 78712. E-mail: dbuss@psy.utexas.edu
DOI: 10.1037/a0021355
Deep-Level Diversity and
Leadership
Kristen M. Klein and Mo Wang
University of Maryland
In the special issue on Diversity and Lead-
ership (April 2010), the authors made a
strong case for the importance of diversity
in workplace leadership, rejected prema-
ture declarations that workplace discrimi-
nation is obsolete, and called for leadership
theories that acknowledge and promote the
value of diversity. We appreciate all au-
thors’ stressing that the glass ceiling still
exists, not only for women but for other
historically low-power groups as well. We
also agree that modern theories of leader-
ship can benefit immensely from increased
participation by scholars and practitioners
who are not Western, White, upper-class
men (Chin, 2010).
In spite of these strengths, we must
admit our surprise at the way in which the
authors of the special issue implicitly de-
fined diversity. Specifically, they focused
primarily on surface-level diversity, or het-
erogeneity in salient, visible characteristics
(e.g., gender, race/ethnicity). However, in
recent years, organizational researchers
have increasingly focused on examining
deep-level diversity, defined as heterogene-
ity in underlying psychological character-
932 December 2010 ●American Psychologist

istics such as personality, attitudes, and
values (Harrison, Price, & Bell, 1998). The
authors’ choice to exclusively address sur-
face-level diversity has several implica-
tions for their discussion of diversity’s role
in leadership.
Specifically, in arguing that surface-
level diversity impacts leadership, it is crit-
ical to disentangle the two leadership-
related variables it supposedly affects:
leadership opportunities and leadership ef-
fectiveness. As to the former, surface-level
characteristics can lead to problematic dis-
criminatory disparities in leadership oppor-
tunities, whereas deep-level characteristics
generally do not; therefore, the authors’
focus on the role of surface-level diversity
in leadership opportunities makes sense.
However, it is not so clear that surface-
level diversity is a major factor in deter-
mining leadership effectiveness, despite
some of the special issue authors’ sugges-
tions to the contrary.
For example, we question Eagly and
Chin’s (2010) argument that people with
certain surface-level characteristics may be
superior leaders, owing to character-build-
ing experiences with discrimination. Al-
though resilient women and minorities may
gain leadership skills from their experi-
ences with discrimination (Szalacha et al.,
2003), other less resilient individuals may
exhibit maladjustment, self-loathing, and
other negative outcomes that are not con-
ducive to effective leadership. Similarly,
we are skeptical about the notion that peo-
ple with certain surface-level characteris-
tics possess special qualities that predis-
pose them to be better leaders than others.
This argument may seem a fair method of
tipping the balance in favor of historically
low-power groups; however, it ultimately
relies on trading out negative stereotypes
for positive ones.
In essence, our primary concern with
the special issue on Diversity and Leader-
ship is that the special issue authors often
conflated surface-level characteristics (e.g.,
gender) with deep-level characteristics
(e.g., expressivity), simply because the two
tend to covary in some instances. However,
these covariations are not sufficient to
claim that surface-level diversity reflects
deep-level characteristics that are critical to
leadership effectiveness. Rather, we argue
that modern leadership scholars could gain
new insights through shifting to a model
that teases the two sources of diversity
apart by emphasizing surface-level diver-
sity in recruitment and selection for lead-
ership positions and deep-level characteris-
tics in leader effectiveness and training.
A substantial body of research on
deep- and surface-level diversity in the
workplace has repeatedly shown that
whereas the negative impacts of surface-
level diversity decrease over time in work
groups, deep-level similarity (e.g., in val-
ues, goal orientations, and personality)
consistently predicts positive workplace
outcomes (e.g., turnover, job attitudes,
team performance; Harrison, Price, Gavin,
& Florey, 2002; Liao, Chuang, & Joshi,
2008; van Emmerik & Brenninkmeijer,
2009). The theoretical implications of these
findings can contribute much to the current
discussion of diversity in leadership.
First, over time, deep-level similari-
ties could override surface-level differ-
ences in determining leader effectiveness
(Hiller & Day, 2003). This suggests that
recruiting and selecting leaders for surface-
level diversity may have few harmful or-
ganizational effects in the long run. More-
over, training leaders to foster shared
organizational values, goals, and attitudes
among their team members could contrib-
ute to deep-level similarity within their
teams, which might result in positive orga-
nizational outcomes over time.
Second, findings suggesting the spe-
cial effectiveness of leaders who possess
cross-category (e.g., androgynous) leader-
ship skills likely reflect the impact of deep-
level qualities on leader effectiveness, and
they suggest that backlash for violating tra-
ditional leader role expectations could be
decreasing in the workplace (see Ayman &
Korabik, 2010). Because of changing soci-
etal norms, younger generations of women
and minority leaders may be internalizing a
broader set of deep-level leadership quali-
ties, free of stereotypical connections with
their surface-level characteristics. As such,
the next generation of leaders may be better
equipped for leadership and more willing
to accept traditional role violations by lead-
ers, particularly if the violator can “do it
all” (e.g., be agentic and expressive).
Finally, by exploring the role of deep-
level characteristics in leadership, re-
searchers and practitioners can identify
fundamental psychological variables that
predict a host of work outcomes, including
leader effectiveness. We suggest that this
endeavor would complement the dispro-
portionate focus to date on surface-level
diversity in leadership, which is not always
useful for predicting and improving lead-
ership ability. For one, prescribed and pro-
scribed roles for groups with distinct sur-
face-level characteristics are too variable
across time and space for any leadership
theory based on these characteristics to be
global or enduring. Identifying deep-level
characteristics that contribute to effective
leadership might provide a theory that re-
searchers can apply universally regardless
of the leader’s gender, skin color, or sexual
orientation. For another, deep-level charac-
teristics are more mutable than surface-
level qualities (Harrison et al., 1998), and
so there is a greater chance of effecting
change in leadership ability by addressing
deep-level characteristics.
We suggest that researchers could bet-
ter predict and increase leader effectiveness
by explicitly addressing deep-level charac-
teristics in theory and practice. By promot-
ing surface-level diversity in leadership op-
portunities and deep-level similarities in
leadership training, it is conceivable that
organizations could counter adverse impact
in leader selection while also improving
organizational outcomes. Given the contin-
ued pervasive absence of women and mi-
norities in leadership roles, researchers and
practitioners have little to lose (and poten-
tially much to gain) by testing these ideas.
REFERENCES
Ayman, R., & Korabik, K. (2010). Leadership:
Why gender and culture matter. American
Psychologist, 65, 157–170. doi:10.1037/
a0018806
Chin, J. L. (2010). Introduction to the special
issue on diversity and leadership. American
Psychologist, 65, 150 –156. doi:10.1037/
a0018716
Eagly, A. H., & Chin, J. L. (2010). Diversity and
leadership in a changing world. American
Psychologist, 65, 216 –224. doi:10.1037/
a0018957
Harrison, D. A., Price, K. H., & Bell, M. P.
(1998). Beyond relational demography: Time
and the effects of surface- and deep-level di-
versity on work group cohesion. Academy of
Management Journal, 41, 96 –107. doi:
10.2307/256901
Harrison, D. A., Price, K. H., Gavin, J. H., &
Florey, A. T. (2002). Time, teams, and task
performance: Changing effects of surface- and
deep-level diversity on group functioning.
Academy of Management Journal, 45, 1029 –
1045. doi:10.2307/3069328
Hiller, N. J., & Day, D. V. (2003). LMX and
teamwork: The challenges and opportunities
of diversity. In G. B. Graen (Ed.), Dealing
with diversity (pp. 29 –58). Charlotte, NC: In-
formation Age.
Liao, H., Chuang, A., & Joshi, A. (2008). Per-
ceived deep-level similarity: Personality ante-
cedents and impact on overall job attitude,
helping, work withdrawal, and turnover. Or-
ganizational Behavior and Human Decision
Processes, 106, 106 –124. doi:10.1016/j.
obhdp.2008.01.002
Szalacha, L. A., Erkut, S., Coll, C. G., Fields,
J. P., Alarco´n, O., & Ceder, I. (2003). Per-
ceived discrimination and resilience. In S. S.
Luthar (Ed.), Resilience and vulnerability:
Adaptation in the context of childhood ad-
versities (pp. 414 – 435). New York, NY:
Cambridge University Press. doi:10.1017/
CBO9780511615788.019
933December 2010 ●American Psychologist

van Emmerik, I. J. H., & Brenninkmeijer, V.
(2009). Deep-level similarity and group social
capital: Associations with team functioning.
Small Group Research, 40, 650 – 669. doi:
10.1177/1046496409346550
Correspondence concerning this comment
should be addressed to Kristen M. Klein, Depart-
ment of Psychology, University of Maryland,
College Park, MD 20742-0001. E-mail: kklein@
psyc.umd.edu
DOI: 10.1037/a0021830
Are Memberships in Race,
Ethnicity, and Gender
Categories Merely Surface
Characteristics?
Alice H. Eagly
Northwestern University
Jean Lau Chin
Adelphi University
Klein and Wang (2010, this issue) offered a
provocative dichotomy between surface-
level and deep-level characteristics and
their relations to leadership. However, their
distinction is far too simple. Contrary to
this uncomplicated dichotomy between
characteristics visible on the surface and
those existing in the psyche, surface-level
characteristics are inherently linked to psy-
chological characteristics. We explored
these interconnections in our earlier article
(Eagly & Chin, April 2010) and distill
these aspects of our analysis of leadership
in this comment.
We acknowledge that some human at-
tributes are easily recognized by sight.
Gender and race are visible to others, and
sometimes ethnicity and disability are vis-
ible as well. The ease of identifying people
by gender and race is apparent in social
cognitive experiments demonstrating the
ubiquity of quick and automatic categori-
zation of people by their sex and race (e.g.,
Ito & Urland, 2003). It is thus not surpris-
ing that psychologists have acknowledged
that these qualities, along with age, are
primary bases of social categorization.
At the same time, gender, race, and
ethnicity have a psychological reality at
deeper levels than the surface of the human
body. One link between visible character-
istics and deep-level traits exists in the
minds of social perceivers. A basic princi-
ple of human judgment, known as corre-
spondent inference, is that people’s internal
characteristics are inferred from their ob-
servable qualities. Observations that people
with certain visible qualities typically en-
gage in certain activities cause observers to
infer matching psychological traits. For ex-
ample, because women disproportionately
engage in care of young children, perceiv-
ers imbue women with caring and nurtur-
ing qualities (Eagly & Steffen, 1984). To
the extent that racial groups are dispropor-
tionately associated with certain occupa-
tions or activities, perceivers infer corre-
sponding psychological traits. As these
beliefs are shared within cultures, stereo-
types form about social groups from these
surface associations and imbue all of the
groups’ members with psychological traits
that are real in the minds of perceivers.
These stereotypes are the root of much
prejudice and discrimination. Members of
social groups encounter discrimination if
their surface-level category memberships
lead people to believe that they do not
“have what it takes” for success in a lead-
ership role. Regardless of individuals’ ac-
tual qualities, attitudes are less favorable
toward those who are stereotypically mis-
matched with the requirements of a leader
role than toward those who are matched.
This less favorable attitude often results in
discrimination, as Klein and Wang (2010)
acknowledged.
Does this stereotyping based on visi-
ble characteristics have diminishing effects
in groups and organizations as time passes,
as claimed by Klein and Wang (2010)? In
general, the answer to this question is no. A
recent meta-analysis of 108 empirical stud-
ies on processes and performance in 10,632
teams found that “cultural diversity is as-
sociated with higher levels of conflict and
less effective communication in teams that
have spent more time together compared
with teams with less tenure” (Stahl,
Maznevski, Voigt, & Jonsen, 2010, p. 702).
Consistent with the classic contact hypoth-
esis (Pettigrew & Tropp, 2006), there are
conditions under which the unfavorable ef-
fects of diversity erode. However, the in-
crease in negativity that is evidently more
typical in team contexts is not surprising
given common beliefs that many such cat-
egories, especially sex and race, have un-
derlying essences formed of fixed, intrinsic
psychological characteristics. In Haslam,
Rothschild, and Ernst’s (2000) research on
40 social categories, female was judged the
most natural, necessary, immutable, dis-
crete, and stable of these social categories,
followed by Asian, male, blind, Black,
White, and Hispanic. The effects of such
essentialist inferences are unlikely to dis-
appear over time.
A second basis of connection between
memberships in social categories and psy-
chological characteristics consists of the
formation of personal identities based on
such group memberships. These identities
represent one’s psychological relationships
to social categories in which one has mem-
bership (e.g., race, social class, religion).
Although the salience of these identities
varies depending on the situation, they are
generally central aspects of people’s self-
definitions. When people associate these
self-categorizations with psychological
characteristics (in the manner that men as-
sociate masculinity with agentic, assertive
traits), they tend to adopt these qualities as
personal standards to guide their behavior.
They attempt to live up to these standards
by performing behaviors congruent with
their identities (e.g., Wood, Christensen,
Hebl, & Rothgerber, 1997) and by select-
ing into congruent social roles (Evans &
Diekman, 2009). Thus, through the media-
tion of social identities, group member-
ships affect behaviors in organizations and
groups, likely including the exercise of
leadership.
A third sense in which surface-level
characteristics link to individual psycho-
logical characteristics derives from the ex-
periences that people have because of these
characteristics. Whether discrimination oc-
curs and how it is manifested depend on
attributes such as race, ethnicity, disability,
gender, and sexual orientation. As we em-
phasized, individuals from racial and eth-
nic minority groups generally have distinc-
tive experiences deriving from their
negotiation of minority and majority cul-
tures. Such experiences can foster the abil-
ity to shift thinking between contexts (Mo-
linsky, 2007) as well as creative cognitive
processes and superior problem-solving
abilities (Leung, Maddux, Galinsky, &
Chiu, 2008). Through such effects, the de-
gree and type of multicultural experience
can influence leaders’ effectiveness.
Yet another route through which sur-
face-level characteristics connect with psy-
chological characteristics is the common
emergence of strength-based and resiliency
models among members of groups who
have been excluded from leadership. As we
argued (Eagly & Chin, 2010), beliefs that
members of one’s group have unique and
effective leadership skills can be a response
to traditional doubts about and resistance to
leadership by members of one’s group. We
thus noted superiority claims concerning
leadership by women, African Americans,
and gay men. Such claims may be viewed
not as scientifically established generaliza-
tions, but as manifestations of the ways that
members of some groups invoke their iden-
tities to confront the challenges of leading
in the face of historical exclusion from
leadership.
934 December 2010 ●American Psychologist