Leadership as a Social Process: Effectively Navigating a Diverse Workplace

Diversity

• Diversity matters because:

  • There may be bottom-line benefits.

  • It’s a present reality.

  • It’s necessary for social equity.

  • References: Ellemers & Rink, 2016; Galinsky et al., 2015; Phillips, 2014

Diversity Efforts Falling Short

• Examples of diversity efforts failing:

  • Facebook can't figure out how to diversify its staff.

  • STEM workforce is no more diverse than 14 years ago.

  • Diversity at Intel is a work in progress.

  • More faculty diversity is not on tenure track.

  • Science's minority talent pool is growing, but draining away.

  • Apple's diversity report reveals problems, risks, and opportunities facing all corporations.

Diversity Efforts Under Attack

• Diversity efforts are under attack, examples:

  • Trump's sweeping new order tries to dismantle DEI in government and the private sector.

  • The war on D.E.I. Heats Up.

  • Trump puts all US government diversity staff on paid leave 'immediately'.

  • Google scraps diversity hiring goals and cites Trump's DEI orders.

Why Diversity Efforts Fail

• One reason diversity efforts fail is because of difficulties in retention:

  • It’s not that we have trouble finding great [Black professionals] who want to come here… It’s getting them to stay.

  • Successful African-American executives indicated that relationship building required special effort to connect with their majority culture peers.

Anxiety, Tension, and Discomfort

• Anxiety, Tension, Discomfort in intergroup dynamics.

  • References: Bergsieker, Shelton & Richeson, 2010; Brescoll & Uhlmann, 2008; Kraus & Keltner, 2013; Kraus & Mendes, 2014; Swencionis & Fiske, 2016; West, Pearson, & Stern, 2014

Perception of Social Groups

• How are different social groups perceived?

  • References: Dupree, Torrez, Obioha & Fiske, 2021, JPSP; Fiske & Dupree, 2014, PNAS; Dupree & Torrez, 2021, JESP; Torrez, Dupree, & Kraus, 2023, JESP; Dupree, 2024, Handbook of Experimental Social Psychology; Morbidelli, Brands, & Dupree, research ongoing

• How do members of advantaged and marginalized groups communicate with each other?

  • References: Dupree & Fiske, 2019, JPSP; Dupree, 2021, Nature Human Behavior; Dupree, 2022, Handbook of Impression Formation; Dupree, 2024, ASQ; Foster-Gimbel & Dupree, under review; Lewis, Dupree, & Hall, in preparation; Dupree, Carillo, Torrez, & Harris, research ongoing

• How to reduce cross-status divides?

  • References: Dupree & Kraus, 2021, PoPS; Dupree & Boykin, 2021, PIBBS; Dupree, 2021, Psych Inquiry; Callaghan, Harouni, Dupree, Kraus, & Richeson, 2021, PNAS; Dupree, 2022, Time Magazine; Dupree, 2021, Discover Magazine

Associations with Social Groups

• White People Associated with High-Status

  • Reference: Dupree, Torrez, Obioha, & Fiske, 2020, Journal of Personality and Social Psychology

• Black People Associated with Low-Status

  • Reference: Dupree, Torrez, Obioha, & Fiske, 2020, Journal of Personality and Social Psychology

Warmth and Competence Perceptions

• Warmth Competence

  • Fiske & Dupree, 2014, Proceedings of the National Academy of Sciences

  • Perceptions of Jobs: Competence Relates to Status

Meta-Stereotypes

• A person’s beliefs regarding the stereotypes that outgroup members hold about own group

  • Reference: Vorauer, Main, & O’Connell, 1998

Meta-Stereotypes Characteristics

• Tend to be negative in content

• Often activated by concerns

• Implications for intergroup contact

  • Less anticipated enjoyment of contact

  • More negative feelings in anticipation of contact

  • Reference: Vorauer, Main, & O’Connell, 1998

Examples of Meta-Stereotypes

• What stereotypes do you think others hold about a group(s) that you identify with?

How Social Groups are Perceived

• How are different social groups perceived?

  • References: Dupree, Torrez, Obioha & Fiske, 2021, JPSP; Fiske & Dupree, 2014, PNAS; Dupree & Torrez, 2021, JESP; Torrez, Dupree, & Kraus, 2023, JESP; Dupree, 2024, Handbook of Experimental Social Psychology; Morbidelli, Brands, & Dupree, research ongoing

• How do members of advantaged and marginalized groups communicate with each other?

  • References: Dupree & Fiske, 2019, JPSP; Dupree, 2021, Nature Human Behavior; Dupree, 2022, Handbook of Impression Formation; Dupree, 2024, ASQ; Foster-Gimbel & Dupree, under review; Lewis, Dupree, & Hall, in preparation; Dupree, Carillo, Torrez, & Harris, research ongoing

• How to reduce cross-status divides?

  • References: Dupree & Kraus, 2021, PoPS; Dupree & Boykin, 2021, PIBBS; Dupree, 2021, Psych Inquiry; Callaghan, Harouni, Dupree, Kraus, & Richeson, 2021, PNAS; Dupree, 2022, Time Magazine; Dupree, 2021, Discover Magazine

Status and Competence

• High status is often associated with high competence, while low status is associated with low competence.

• Liberals

White Liberals and Stereotypes

• How Do White Liberals Affiliate?

• Stereotypes are:

  • Deeply ingrained

  • Widely known

  • Implicitly held

• Liberal

Competence Downshift

• Competence: White liberals, in seeking to affiliate with Black Americans, draw on stereotypes when responding to a Black (versus White) interaction partner

  • Reference: Dupree & Fiske, 2019, Journal of Personality and Social Psychology

Competence of Words Selected

• Partner Race x Conservatism $\beta = .12$ , $t(425) = 2.56$, $p = .011$

• Liberal $p = .147$ $p = .030$

  • Dupree & Fiske, 2019, Journal of Personality and Social Psychology

Competence Downshift in the Real World

• Competence Downshift in the Real World

Archival Study

• Democratic and Republican presidential candidates

  • Speeches delivered to minority audiences

  • Speeches delivered to mostly-White audiences

Analyzing Campaign Speeches

• Use LIWC software1: Linguistic Inquiry and Word Count

• Create dictionaries2

  • Competence

  • Warmth

• LIWC counts words in each speech

• Dependent variable: % of words in speeches from each dictionary

  • 1 Pennebaker, Booth, & Francis, 2007 | 2 Bartz & Lydon, 2004 | 2 Gaucher, Friesen, & Kay, 2011 | 2 Rudman & Killianski, 2000

Archival Study Results

• Audience Race x Speaker Party $F(1,70) = 0.37$, $p = .547$, $\eta_p{}^2 = .005$

  • Warmth Word Usage (z)

  • Reference: Dupree & Fiske, 2019, Journal of Personality and Social Psychology

Archival Study Results - Competence

• Audience Race x Speaker Party $F(1,70) = 0.04$, $p = .848$, $\eta_p{}^2 = .001$

  • $p = .020$ $p = .110$

  • Competence Word Usage (z)

  • Reference: Dupree & Fiske, 2019, Journal of Personality and Social Psychology

Internal Meta-Analyses

• Results for fixed effects models shown. For all effects sizes, dependent variables were residualized and continuous variables standardized.

• For all effects of Partner Race and interactions with Partner Race, White partner is coded as−1. †p<.10. *p<.05. **p<.01. ***p<.001.

• Indicates the number of studies included in meta-analyses

• Reference: Dupree & Fiske, 2019, Journal of Personality and Social Psychology

Competence Downshift Summary

• White liberals present less competence to Black audiences than White ones

• Is it reversible?

  • Challenge the stereotype

  • Remove the need to prove goodwill

• Is there real-world evidence?

  • White Democratic candidates downshift competence to mostly-minority (versus mostly-White) audiences

  • The Competence Downshift

  • Reference: Dupree & Fiske, 2019, Journal of Personality and Social Psychology

Implications of Competence Downshift

• How

  • Endearing!

• Competence

  • Dupree, Carillo, Torrez, & Harris, research ongoing

Implications of Competence Downshift - Demeaning

• How

  • Demeaning!

• Competence

  • Dupree, Carillo, Torrez, & Harris, research ongoing

Racial Minorities’ Self-Presentation

• Conservative Competence

  • Reference: Dupree, 2021, Nature Human Behavior

Racial Minorities Upshift Competence

• Black and Latinx conservatives upshift competence relative to liberals in mostly white settings

  • Reference: Dupree, 2021, Nature Human Behavior

Women's Self-Presentation

• Dupree, 2024, Administrative Science Quarterly

Stereotypes of Women

• “We hold ourselves back in ways both big and small, by lacking self- confidence, by not raising our hands, and by pulling back when we should be leaning in. We internalize the negative messages we get throughout our lives—the messages that say it’s wrong to be outspoken, aggressive, more powerful than men.”

Stereotypes

• Low competence

• High competence

  • Brands & Kilduff, 2014| DiTomaso et al., 2007 | Eagly & Wood, 2011 | Heilman, 2001 | Rudman et al., 2008 | Ridgeway, 1989

Stereotypes Reinforced

• Stereotypes are:

  • Reinforced by and reinforce social structures

  • Widely known

  • Can à concern

• Low competence

• High competence

  • Brands & Mehra, 2019| Cortland & Kinias, 2019 | DiTomaso et al., 2007 | Fiske, Cuddy, & Glick, 2007 | Hoyt & Murphy, 2016 Kanter, 1977 |Purdie-Vaughs, Steele et al., 2008 | Yan et al., 2021

Women Leaders Reverse Stereotypes

• Do Women Leaders Reverse Stereotypes?

• Low competence

• High competence

Stereotype Reversal in Intergroup Settings

• Reference: Dupree & Fiske, 2019, JPSP | Dupree, Torrez, Obioha & Fiske, 2021, JPSP | Dupree, 2021, Nature Human Behavior Dupree, 2022, Handbook of Impression Formation | Lewis, Dupree, & Hall, in preparation | Foster-Gimbel & Dupree, under review

• Competence, Competence

• High competence, Low competence

Competence Upshift

• Competence: Women (versus men) leaders draw on stereotypes when conversing in mostly-male settings

  • A Competence Upshift

  • Dupree, 2024, Administrative Science Quarterly

• Low competence

• High competence

How Women Upshift Competence

• Often examined unidimensionally

  • Dominance: Influence over others

    • E.g., “determined“, “assertive“

  • Ability: Capability to pursue goals

    • E.g., “educated“, “intelligent“

  • Status: Social standing, reputation

    • E.g., “influential“, “superior“

  • Abele et al., 2016 | Anderson, Hildreth, & Howland, 2015 | Dupree et al., 2021 | Kelter et al., 2008 | Koenig & Eagly, 2011 | Ma, Rosette, Koval, & Livingston, 2015 | Zou & Cheryan, 2017

Leadership and Dominance

• Being stereotyped as submissive is a particular burden for leaders.

• Inferred dominance à election outcomes

  • United States

  • France

  • Switzerland

• Westerners prefer leaders dominant.

  • Antonakis & Dalgas, 2009 | Ballew & Todorov, 2007 | Klofstad, 2016 | Todorov et al, 2005

Dominance Upshift

• Dominance: Women (versus men) leaders draw on stereotypes when conversing in mostly-male settings

  • A Dominance Upshift

  • Dupree, 2024, Administrative Science Quarterly

• Low dominance

• High dominance

Women Upshift Dominance

• Intersectionality

  • People who share one demographic category (e.g., gender) but differ on another (e.g., race) experience different outcomes

• Intersectional analyses = necessary

  • To determine which women reverse submissiveness stereotypes by using more dominant language than men

  • Cole, 2019 | Collins, 2000 | Crenshaw, 1991 | Hall, Hall, & Galinsky, 2019 | Remedios & Snyder, 2015

Women Upshift Dominance - Gender Stereotypes

• Gender stereotypes are racialized.

  • White women?

    • “submissive“, “communal“

  • Black women?

    • “angry“, “masculine“

  • Latina women?

    • “feisty“, “hot-blooded“

  • Brescoll, 2015 |Collins, 2000 | Crenshaw, 1991 |Goff, Thomas, & Jackson, 2008 | Hall et al., 2012, 2019 | Rosette et al., 2016 | Williams, Phillips, & Hall, 2016

Gender Stereotypes

• Gender stereotypes are racialized.

  • White women? (H1a)

    • “submissive“, “communal“

  • Black women? (H1b)

    • “angry“, “masculine“

  • Latina women? (H1b)

    • “feisty“, “hot-blooded“

  • Collins, 2000 | Crenshaw, 1991 |Goff, Thomas, & Jackson, 2008 | Hall et al., 2012, 2019 | Rosette et al., 2016 | Williams, Phillips, & Hall, 2016

Women's Race

• Women's Race

• Women Leaders Counter Stereotypes by Using Dominant Language

• Gender Stereotypes Depict Women as Low in Dominance

• Constituents Reject Women Leaders

• Women's Race

• Journalists Depict Women Leaders as Dominant but Cold

  • Dupree, 2024, Administrative Science Quarterly

Congressional Speech

• Congressional Speech

Speech Collection

• Black and Latinx Senators and Representatives ~ 250,000+ remarks delivered at Senate or House

• Paired with White Senators and Representatives

Speech Analysis

• SADCAT: Semi-Automated Dictionary Creation for Analyzing Text

• Select dictionary

  • High power

• SADCAT counts words in each politician’s aggregated remarks

• Dependent variable: % of words in remarks from high power dictionary

  • Nicolas, Bai, & Fiske, 2022

Congressional Speech Analysis

• 310 Black, Latinx, and White lawmakers (90 women)

  • 250,000 Congressional remarks

  • Analyzed stereotype content of remarks using SADCAT

  • Controlled for legislator party, ideology, congressional experience

  • p < .001 $p = .056$

Results of Speech Analysis

• Table 2. Effect of Legislator Gender Predicting Legislators' Use of Dominant Language in
  Congress (Study 1; Linear Regression Models, N = 310)ª

• a Unstandardized regression coefficients are presented; standard errors are in parentheses.

  • Dupree, in press, Administrative Science Quarterly

Results of Speech Analysis - Figures

• Female politicans Male politicians

  • $p = .013$ $p = .189$

• Female politicans Male politicians

• $p = .046$

• Dominant Language in Congressional Speech (z)

  • Dupree, in press, Administrative Science Quarterly

Congressional Tweets

• Congressional Tweets

Tweet Collection

• Black, Latinx, White lawmakers in 116th Congress

• Twitter API scraped tweets by each lawmaker

  • ~ One million tweets

Tweet Analysis

• SADCAT: Semi-Automated Dictionary Creation for Analyzing Text

• Select dictionary

  • High power

• SADCAT counts words in aggregated tweets

• Dependent variable: % of words in tweets from high power dictionary

  • Nicolas, Bai, & Fiske, 2022

Congressional Tweets (Study 2)

• 511 Black, Latinx, White and lawmakers (116 women)

  • One million tweets

  • Analyzed stereotype content of tweets using SADCAT

  • Controlled for legislator party, ideology, congressional experience

Results of Tweet Analysis

• Table 4. Effect of Legislator Gender And Race Predicting Legislators' Use of Dominant
  Language on Twitter (Study 2; Linear Regression Models, N = 511)ª

• a Unstandardized regression coefficients are presented; standard errors are in parentheses.

  • Dupree, in press, Administrative Science Quarterly

Tweet Results - Figures

• Female politicans Male politicians

  • p < .001 $p = .510$

• Female politicans Male politicians

  • $p = .008$

• Dominant Language in Congressional Tweets (z)

  • Dupree, 2024, Administrative Science Quarterly

Backlash Effect

• Women are penalized for dominant behavior.

  • Women who express dominance…

  • Deemed less likeable, less hireable

  • Conferred lower status, offered lower salaries

  • Several unknowns.

    • Do women leaders experience backlash?

    • Does use of dominant language à backlash?

  • Dominance

  • Heilman & Okimoto, 2007 | LaFrance, 1992 | Okimoto & Brescoll, 2010 | Rosette & Tost, 2010 | Rudman, 1998 Rudman & Phelan, 2008 | Williams & Tiedens, 2016

Media Portrayals

• Media Portrayals

Editorial Language Use

• Dominance

• Women leaders’ use of dominant language at work predicts journalists’ use of dominant but cold language in editorials

  • Study 3: Media Portrayals

  • Dupree, 2024, Administrative Science Quarterly

Intersectionality Importance

• Black and Latina women may be more or less likely to experience backlash for dominant language.

  • Intersectional invisibility

    • Women of color à less discrimination due to relative invisibility, non-prototypicality

    • à predicted effects attenuated (H3a)

  • Double jeopardy hypothesis

    • Women of color à more discrimination due to racism and sexism

    • à predicted effects exacerbated (H3b)

  • Beale, 1970 | Berdahl & Moore, 2006 |Chow, 1987 | Epstein, 1973 | hooks, 1989 | Purdie-Vaugns & Eibach, 2008 |Reid, 1984 | Rosette & Livingston, 2012

Perceived Backlash

• Imagine yourself in a high-profile leadership position at a majority-male workplace. You are one of the first and only [Black] [Latina] women in such a high- ranking leadership position. As part of your role, you regularly lead meetings, address the general public, and post on social media. How would others react if you presented yourself as…

  • 277 employed Black, Latina, and White women

  • $p = .048$ $p = .048$

  • Dupree, 2024, Administrative Science Quarterly

Editorial Collection

• Factiva to scrape written editorials about politicians analyzed in Study 1

  • ~ 18,000 editorials

Analysis of Editorials

• SADCAT : Semi-Automated Dictionary Creation for Analyzing Text

• Select dictionaries

  • High power

• Low sociability

• SADCAT counts words in aggregated editorials

• Dependent variable: % of words in editorials from high power, low sociability dictionaries

  • Nicolas, Bai, & Fiske, 2022

Media Portrayals Analysis

• 274 leaders (77 women, 88% Study 1 sample)

  • 18,000 editorials

  • Analyzed stereotype content of editorials using SADCAT

  • Controlled for legislator party, ideology, congressional experience

  • p < .001 $p = .056$

Results of Media Portrayals

• Table 6. Effect of Legislator Gender, Race, and Use of Dominant Language in Congress
  Predicting Journalists' Use of Dominant Language in Editorials (Study 3; Linear Regression
  Models, N=287)ª

• Table 7. Effect of Legislator Gender, Race, and Use of Dominant Language in Congress Predicting Journalists' Use of Cold Language in Editorials (Study 3; Linear Regression Models, N = 287)

  • Dupree, in press, Administrative Science Quarterly

Results of Media Portrayals - Figures

• Female politicians Male politicians

  • $p = .001$ $p = .229$

• Female politicians Male politicians

  • $p = .021$ $p = .972$

• Dominant Language in Editorials (z), Dominant Language in Congressional Speech (z)

• Cold Language in Editorials (z), Dominant Language in Congressional Speech (z)

  • Dupree, in press, Administrative Science Quarterly

Media Portrayals - Figures Cont.

• Black and Latinx politicians

  • p < .001 $p = .533$

• Female politicians Male politicians

  • p < .001 $p = .816$

  • Dupree, 2024, Administrative Science Quarterly

Constituent Reactions

• Constituent Reactions

Constituent Reactions Analysis

• 794 working adults (352 women, 94% registered voters)

  • Simulated social media profile

  • 2 (Dominance References) x 2 (Leader Gender) x 2 (Leader Race)

  • Likeability ($\alpha = .85$), Leadership Suitability ($\alpha = .93$), Dominance ($\alpha = .90$), Warmth ($\alpha = .93$), Competence ($\alpha = .89$)

  • Brescoll, 2011 | Okimoto & Brescoll, 2012 | Rudman et al., 2012

Gender Manipulation

• Gender Manipulation

Dominance Manipulation

• Dominance Manipulation

Constituent Reactions - Results

• White Leaders Dominance References x Leader Gender $F(1,387) = 0.84$, $p = .360$

  • $p = .269$ $p = .014$

  • Dupree, in press, Administrative Science Quarterly

Black Leaders - Results

• Black Leaders Dominance References x Leader Gender $F(1,399) = 3.87$, $p = .049$

  • $p = .007$ $p = .792$

  • Dupree, in press, Administrative Science Quarterly

How Groups are Perceived

• How are different social groups perceived?

  • References: Dupree, Torrez, Obioha & Fiske, 2021, JPSP| Fiske & Dupree, 2014, PNAS | Dupree & Torrez, 2021, JESP | Torrez, Dupree, & Kraus, 2023, JESP Dupree, 2024, Handbook of Experimental Social Psychology | Morbidelli, Brands, & Dupree, research ongoing

• How do members of advantaged and marginalized groups communicate with each other?

  • References: Dupree & Fiske, 2019, JPSP| Dupree, 2021, Nature Human Behavior | Dupree, 2022, Handbook of Impression Formation | Dupree, 2024, ASQ Foster-Gimbel & Dupree, under review | Lewis, Dupree, & Hall, in preparation | Dupree, Carillo, Torrez, & Harris, research ongoing

• How to reduce cross-status divides?

  • References: Dupree & Kraus, 2021, PoPS | Dupree & Boykin, 2021, PIBBS | Dupree, 2021, Psych Inquiry | Callaghan, Harouni, Dupree, Kraus, & Richeson, 2021, PNAS Dupree, 2022, Time Magazine | Dupree, 2021, Discover Magazine

Egalitarian Norms

• Egalitarian Norms

• Gracism

• Eracism t-shirt

• Sinclair et al, 2006

• Blank t-shirt

Friendly vs Rude Behavior

• Behavior

• Friendly Rude

• Implicit Anti-Black Prejudice

• Egalitarian Neutral

• Sinclair et al, 2006

Insights for Organizations

• Stereotypes continue to shape hiring and policy preferences

  • Race-status associations

  • Actively counter narratives, diversifying employees across hierarchy

• Allies may contribute to the problem too

  • Competence downshift

  • Guided onboarding, reshaping diversity trainings

• Need to acknowledge intersectionality

  • Stereotypes, experiences, and behaviors of women differ by race

  • More intersectional scholarship and organizational initiatives

  • Dupree & Fiske, 2019, JPSP | Dupree, Torrez, Obioha, & Fiske, 2021, JPSP | Dupree, 2021, Nature Human Behavior Dupree, 2022, Handbook of Impression Formation |Dupree, 2024, ASQ

Leadership is a Social Process

• Leadership is a social process!

• To be an effective leader in an increasingly diverse world, must navigate diverse workforces

• Barriers before, during, and after intergroup

  • à Anticipation of negative contact

  • à Divergent goals and behaviors during contact

  • à Backlash encountered after contact

• Traversing barriers

  • By affirming counter-stereotypes

  • By promoting egalitarian norms, behaviors

Individual Steps

• What can YOU do?

• What steps can you take to lead organizational change surrounding diversity-related outcomes?

  • Individually

  • In social/professional circles

  • In organizations