Best Practices or Best Guesses How to Increase Diversity in Corporate Leadership

1
Best Practices or Best Guesses?How to Increase
Diversity in Corporate Leadership

• Alexandra Kalev
• University of Arizona, Department of Sociology

NCWIT Seattle, WA, November 6-7, 2007
Thanks to the National Science Foundation, the
Russell Sage Foundation, the Robert Wood Johnson
Foundation, and to my collaborators on various
parts of the research Frank Dobbin and Erin
Kelly.
2
How Should We Judge Diversity Programs?

• Self-reporting of Attitudes
• Workplace Culture Survey
• Employee Scores on Bias Index
• Companys Productivity or Performance
• Diversity of Workers
• Diversity of Managers

3
The Sum Total of What We Know AboutAffirmative
Action Diversity Management
Note It Fits on One Slide

• 1970s Federal Contractors Hired More Black Men
• 1980s Federal Contractors Same as Others
• 1975-76 Compliance Reviews Helped, Affirmative
  Action Plans Helped

4
What Experts Prescribe

• Combat Psychological Bias
• Oversight and Feedback gt Diversity Evaluations
• Consciousness gt Diversity Training
• Combat Social Isolation
• Networking
• Mentoring
• Establish Responsibility (Organizational
  sociology)
• Diversity Committee
• Full Time Staff
• Benchmarking plan

5
National Sample of 829 Firms in 2002
6
Meanwhile Unrelated Developments

• Improving efficiency and quality / Employee
  Empowerment Programs
• Self-directed work teams
• Cross-training/Job rotation
• gtgt Greater visibility and exposure for women and
  minorities

7
(No Transcript)
8
Data (I)

• Life Histories of more than 800 Organizations
  from 1971-2002
• EEO1 Establishment Reports 1971-2002
• Race, Gender, Ethnicity by 9 Occupational
  Categories
• Population Employers with 100 or more workers,
  Federal Contractors with 50 or more
• 250,000 workplaces in 2002

9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
Data (II)

• 2002 Retrospective Survey of Workplaces
• Sampled from 1999 EEOC Dataset
• Stratified by Industry, Size, and Age in EEOC
  Dataset
• Response Rate of 67
• N829

13
(No Transcript)
14
Method

• Pooled Cross-sectional Time-Series Models
• Fixed Effects for Workplace and Year
• Seemingly Unrelated Regression (error terms are
  correlated, SUR produces unbiased and efficient
  estimates).
• Robust to Cochrane-Orcutt Correction for Serial
  Correlation.
• Modeling log odds of Proportion of Group in Mgt.
  (logged to predict growth at low end of range)
• Controls for everything under the sun.
• Effect of New Policy on Representation of Each
  Group among Managers.

15

• FINDINGS (I)

16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
What Have We Learned So Far (I)

• Most effective Establishing organizational
  responsibility for change mentoring
• Least effective Diversity training, evaluations
  and networking programs
• May even backfire
• The surprise of the ball Team work and cross
  training help women and minorities get ahead gtgt
  let them shine

22
What Have We Learned So Far (II)

• Diversity Begets Diversity Once in leadership
  positions, women and minorities help create more
  diversity

23
Diversity Training -Commitment and Resistance
24
What We Know About Diversity Training

• Diversity training has no positive impact on
  workplace demographics on average.
• Mixed evidence about the effect of diversity
  training on managerial attitudes.
• Diversity training can create resistance.

25
ParticipationWhat Experts Prescribe

• Economics, Management Theory
• Mandates incentives promote effort
• Social Psychologists
• Cognitive dissonance Choice increases commitment
  
• Inducement reduces motivation

26
Content Threat vs. Business Case

• Forbidden Toy Weak sanction increase
  compliance
• Fines or rewards undermine internal motivation at
  work
• Diversity training that focuses on legal
  requirements leads to more negative responses

27
Hypotheses

• Voluntary diversity training is more effective
  than mandatory
• Focus on organizational motives for increasing
  diversity is more effective than legal motives

28

• FINDINGS (II)

29
(No Transcript)
30
(No Transcript)
31
What Should Employers Do? (I)

• Assign Responsibility Create a Task Force, Hire
  a Diversity Manager
• This Has a Second Order Effect Effectiveness of
  other Programs is Increased
• Evidence-Based Management Dont be afraid of
  numbers - Use Benchmarks
• Craft Training Carefully, Making it Voluntary,
  and Focusing on Cultural Awareness
• Set up Mentoring Programs

32
What Should Employers Do? (II)

• Let them Shine
• Segregated jobs and narrow specialization
  disfavor women and minorities. Teams and cross
  training increase efficiency and equality.

33

• THIS IS THE END

34

• Questions? Comments? Ideas?
• akalev_at_email.arizona.edu