Normative messages and students’ intentions to engage in collaborative learning: evidence from an experimental study

Introduction Despite the recognized benefits of collaborative learning, students’ voluntary participation in activities such as group reading remains limited. This study examined students’ intentions to engage in collaborative learning and explored whether a brief normative-message intervention varied as a function of norm framing and message source. Methods In a 2 × 2 between-subjects online experiment, 183 students at the University of Tehran were randomly assigned to one of four conditions combining dynamic versus static descriptive norms with identity-proximal versus identity-distal message sources. Subjective behavioral intention and anticipated participation frequency were analyzed as separate outcomes. Attitudes, injunctive and descriptive norms, perceived behavioral control, behavior-related student identity linkage, and previous-semester participation frequency were also assessed. Data were analyzed using multiple regression, two-way multivariate analysis of variance, and exploratory moderation analyses. Results Attitudes, perceived behavioral control, previous-semester participation frequency, and behavior-related student identity linkage were significant correlates of anticipated participation frequency, whereas demographic characteristics, injunctive norms, and descriptive norms were not significant predictors. However, the experimental manipulation was unsuccessful: participants receiving dynamic and static messages did not differ significantly in perceived past-to-present norm change or expected future norm growth. Consistent with the failed manipulation, neither assigned norm type nor assigned message source, nor their interaction, significantly affected either outcome, and effect sizes were negligible. An exploratory interaction between assigned norm condition and behavior-related student identity linkage reached statistical significance, but no conditional effect was significant and no region of significance was identified; therefore, this finding should not be interpreted as evidence of a substantive moderation effect. Discussion Because perceived source proximity was not assessed following the manipulation, the null source effect cannot be interpreted as evidence that source proximity is ineffective. Overall, the study documents an unsuccessful single-exposure normative-message manipulation while identifying attitudinal, perceived-control, prior-participation, and identity-related variables as correlates of participation intentions. The findings underscore the importance of verifying normative perceptions and source identification before drawing conclusions about the downstream effects of normative interventions.