Latent profile analysis as a method for process evaluations: Discovering response subgroups in a mindfulness...

Area of Science:

• Psychology
• Intervention Science
• Adolescent Health

Background:

• Clustering techniques are used in intervention studies to identify participant subgroups.
• However, their application in understanding diverse motivational and behavioral responses to new behaviors is underexplored.

Purpose of the Study:

• To utilize latent profile analysis to identify distinct social cognitive response types to a mindfulness intervention.
• To identify longitudinal classes of mindfulness practice behavior trajectories.

Main Methods:

• Employed latent profile analysis on data from a school-based mindfulness intervention (N=1646) with 12-15 year olds in Finland.
• Explored associations between identified classes, gender, linguocultural group, mental health, and practice outcomes.

Main Results:

• Identified five social cognitive classes (e.g., Uncertain but Positive, Disinclined) and five practice trajectory classes (e.g., Stable Low, Increased from Zero).
• Descriptive norm was the strongest differentiator for social cognitive classes.
• Found associations between classes and linguocultural groups and mental health, but not gender.

Conclusions:

• Person-centered analyses offer valuable insights for intervention process evaluations.
• Findings suggest tailoring universal interventions to specific subgroup profiles can enhance acceptability and engagement.