Hilliard and Liben (2010)


• Aim: 



  To investigate how social category salience affects the development of gender stereotypes and inter-group behavior in preschool children.
• Method


  • Design: Field experiment with pre-test/post-test design
  • Sample: 57 US children (age range: 3 years 1 month to 5 years 6 months)
• IV & DV: 


  • Independent Variable: Gender salience (high vs. low)
  • Dependent Variables:

  1. Gender attitude (measured by POAT-AM test)
  2. Play behavior with opposite-sex peers
• Procedure



  1. Pre-test: Children completed POAT-AM gender attitude test
  2. Observation of play behavior
  3. Two-week intervention:

  • High salience condition: Gender-specific language and organization
  • Low salience condition (control): No changes to usual practices

  1. Post-test: Repeat of POAT-AM and play behavior observation
  2. Debriefing program to counteract potential negative effects
• Results


  1. High salience condition:

  • Significant decrease in "both" responses on POAT-AM (increased gender stereotypes)
  • Significant decrease in play with opposite-sex peers

  1. Low salience condition:

  • No significant changes in gender attitudes or play behavior
• Conclusion: 


  Increased gender salience leads to increased gender stereotypes and decreased inter-group play in preschool children, supporting Social Identity Theory's predictions about in-group favoritism and out-group homogeneity.
• Strengths:


  1. High ecological validity due to the naturalistic setting (field experiment)
  2. Experimental design allowing for cause-effect inferences
  3. Use of both attitudinal (POAT-AM) and behavioral (play observation) measures
  4. Inclusion of a debriefing program to address ethical concerns
  5. Pre-test/post-test design allows for clear comparison of effects
• Limitations:


  1. Low internal validity due to less controlled environment
  2. Sampling bias: Middle to upper-class children (non-free preschool). Preschools with gender-neutral policies (potentially biased sample)
  3. Difficulty in measuring exact level of gender salience
  4. Ethical concerns about potential lasting effects on children's behavior
  5. Limited generalizability due to specific sample characteristics