Group cohesion in older adult exercisers: Prediction and intervention effects

Paul A. Estabrooks, Albert V. Carron

Research output: Contribution to journalArticlepeer-review

117 Scopus citations


Two studies were conducted to examine the relationship between class cohesion and exercise adherence in older adult exercisers. Study 1 examined the predictive ability of four dimensions of cohesion on exercise participation at 1, 6, and 12 months following the initial assessment of cohesion. Study 2 examined the effectiveness of a team-building intervention, designed to enhance class cohesion (and based on Study 1 results), on improving exercise adherence and return rates. Participants were assigned to a team-building, placebo, or control condition. Study 1 showed that three measures of cohesion. Individual attractions to the group-social, Group integration-social, and Group integration-task, were all significantly related to exercise class attendance following a 1-month interval. Group integration-task was significantly related to class attendance following a 6- and a 12-month interval. Study 2 showed that participants in the team- building condition (a) attended more classes than the control and placebo conditions and (b) had a higher return rate following a 10-week hiatus than the control condition. It was concluded that (a) class cohesion plays a significant role in exercise class participation, both short- and long-term, and (b) samples of older adult exercisers are appropriate groups for interventions based on developing class cohesion.

Original languageEnglish (US)
Pages (from-to)575-588
Number of pages14
JournalJournal of Behavioral Medicine
Issue number6
StatePublished - 1999
Externally publishedYes


  • Adherence
  • Attractions to the group
  • Group integration
  • Team-building

ASJC Scopus subject areas

  • General Psychology
  • Psychiatry and Mental health


Dive into the research topics of 'Group cohesion in older adult exercisers: Prediction and intervention effects'. Together they form a unique fingerprint.

Cite this