Streamlining mindfulness assessment in young adults: a machine learning approach using CAMS-R

This study validates the effectiveness of streamlining the Cognitive and Affective Mindfulness Scale-Revised (CAMS-R) using supervised machine learning, significantly enhancing the efficiency of mindfulness assessments for young adults. Traditional mindfulness scales, while reliable, often demand considerable time and cognitive effort from respondents, particularly in large-scale studies. By applying advanced machine learning models, such as Logistic Regression and Support Vector Machines (SVM), we identified the key questions that best predict mindfulness levels, enabling a more refined and efficient scale without compromising its validity. The machine learning models consistently emphasized the importance of questions related to non-judgmental observation and present-moment awareness, confirming these dimensions as essential for accurately measuring mindfulness. Logistic Regression, which achieved a high accuracy rate of 93.2%, emerged as the most effective model, closely followed by SVM. By focusing on the most predictive items, this data-driven approach reduces the cognitive load on respondents, while simultaneously enhancing the quality of data collected. The strong agreement between models in identifying critical questions further supports the validity of this approach. Streamlining the questionnaire through machine learning not only increases efficiency but also preserves the psychometric integrity of the CAMS-R, ensuring that it remains a reliable and valid tool across various contexts. This refined assessment tool is particularly valuable in educational and clinical settings, where both time and precision are critical. Therefore, employing machine learning to optimize mindfulness assessments represents a significant advancement in psychological measurement, combining ease of use with scientific rigor.