When surface and deep-level diversity collide: The effects on dissenting group members

Katherine W. Phillips, Denise Lewin Loyd

Research output: Contribution to journalArticlepeer-review


Diversity researchers have distinguished between surface-level (e.g., social categories) and deep-level (e.g., attitudes, opinions, information, and values) diversity, but have not fully explored the complexities of their simultaneous existence in groups. We examined how the relationship between surface-level and deep-level diversity impacts the emotional and behavioral reactions of dissenting group members and the effectiveness of decision-making groups. We conducted two studies focusing on dissenting social majority members (individuals who hold dissenting deep-level task perspectives yet belong to the surface-level majority) in three-person groups. The results show that surface-level diverse groups (with two similar and one dissimilar individuals) were perceived as more positive and accepting, fostered more persistent and confident voicing of dissenting perspectives, and displayed greater task engagement than surface-level homogeneous groups (containing all similar individuals). Surface-level diversity (both task-relevant and irrelevant) may be beneficial for groups even when the group member who is different on the surface does not have a different deep-level task perspective to share. We discuss implications for understanding how surface-level diversity affects organizational work groups.

Original languageEnglish (US)
Pages (from-to)143-160
Number of pages18
JournalOrganizational Behavior and Human Decision Processes
Issue number2
StatePublished - Mar 2006
Externally publishedYes


  • Dissent
  • Diversity
  • Expectations of similarity
  • Social categorization

ASJC Scopus subject areas

  • Applied Psychology
  • Organizational Behavior and Human Resource Management


Dive into the research topics of 'When surface and deep-level diversity collide: The effects on dissenting group members'. Together they form a unique fingerprint.

Cite this