The longitudinal links between shame and increasing hostility during adolescence

@article{heaven2009the,
  author = {Heaven, Patrick C.L. and Ciarrochi, Joseph and Leeson, Peter},
  title = {The longitudinal links between shame and increasing hostility during adolescence},
  year = {2009},
  journal = {Personality and Individual Differences},
  volume = {47},
  number = {8},
  pages = {841-844},
  publisher = {Elsevier BV},
  doi = {10.1016/j.paid.2009.07.002}
}

In plain language

Shame is often used as a tool to push people toward better behavior — think of the phrase "you should be ashamed of yourself." But does shaming actually work, or does it backfire? Earlier studies had shown that shame and hostility tend to go together, but because those studies were snapshots in time, no one knew which came first. This study followed the same adolescents over a full year to find out.

As part of the Wollongong Youth Study in Australia, 765 grade 9 students (average age about 14) reported how often they had felt shame-related emotions (like ashamed, disgusted with self, dissatisfied with self) and hostile emotions (like angry, hostile, irritable). A year later, 670 of them completed the same measures in grade 10. Statistical modeling then tested whether shame led to hostility, hostility led to shame, or the two fed each other.

The results clearly favored the shame-first model. Teens who felt more shame in grade 9 became more hostile by grade 10, even after accounting for how hostile they already were — and this held for both boys and girls. Hostility, by contrast, did not predict later increases in shame. The findings suggest shame has a paradoxical effect: rather than reshaping behavior in a positive direction, it appears to make aggressive, antisocial responses more likely.

Key findings

• Shame and hostility were both quite stable over one year, and strongly correlated with each other at each time point.
• Higher shame in grade 9 predicted increases in hostility in grade 10, after controlling for grade 9 hostility — supporting the shame-antecedent model.
• Hostility did not predict increases in shame, ruling out the shame-as-consequence and reciprocal-influence models.
• The effect of shame on rising hostility held for both girls (beta = .18) and boys (beta = .20).
• Feeling guilt, unlike the five shame items, did not significantly predict later hostility — supporting the distinction between shame (a negative judgment of the whole self) and guilt (a judgment of a specific behavior).
• The authors conclude that shame is not an effective way to shape prosocial behavior and may instead provoke antisocial responses.

Related work

• All publications by Joseph Ciarrochi (searchable, with free PDFs)
• Process-Based Therapy & Idionomic Analysis

Author: Joseph Ciarrochi (ORCID 0000-0003-0471-8100). Free copy hosted with permission for scholarly use. Please cite the published version via the DOI above.