Poster #211 - Measuring Adolescents’ Mindfulness: Psychometric Properties of a 20-item Five Facet Mindfulness Questionnaire

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Social, Emotional, Personality

Integrative Statement

Background: Mindfulness-based interventions are now routinely being used in schools and clinics around the world to help adolescents cope with stress (Abujaradeh, Safadi, Sereika, Kahle, & Cohen, 2018), and even do better in school (Lin & Mai, 2018). Beyond interventions, there is also a growing appreciation for the need to understand naturalistic developmental changes in mindfulness across adolescence (Roeser & Eccles, 2015), and how such changes may relate to positive developmental outcomes. To accomplish these goals, the field needs valid and reliable measures of mindfulness. Though measures for assessing mindfulness in adolescents exist, several shortcomings (e.g., focus on individual rather than multiple mindfulness facets) prompted us to test the psychometric properties of one of the most widely-used mindfulness self-report instruments with adults, the Five Facet Mindfulness Questionnaire (FFMQ). The FFMQ and its various short-forms do reflect the multifaceted nature of mindfulness, but have never been validated for use with adolescents living in the United States. Thus, it is unknown whether this instrument can be productively used in mindfulness studies with adolescents.

Objective: To examine the reliability, factor structure, measurement invariance, and construct validity of a short-form (20-item) version of the FFMQ in a sample of North American adolescents.

Methods: 599 high school students (Mean age=16.27, SD=1.15; 49% female) completed a 20-item version of the FFMQ at three assessment waves, spaced roughly 3 months apart, over the course of one academic year. They also completed other measures of social-emotional functioning. Confirmatory factor analyses (CFA) tested the theorized factor structure and measurement invariance (across gender, grade level, and assessment wave). Convergent validity was assessed by examining bivariate correlations between the FFMQ and theoretically-related constructs, including self-compassion and perceived stress.

Results: FFMQ facets showed adequate reliability at wave 1 (Cronbach’s alphas =.61 to .88), wave 2 (Cronbach’s alphas =.61 to .86 ), and wave 3 (Cronbach’s alphas=.67 to .88). Consistent with prior research, CFA showed that a four-factor, hierarchical model (excluding the observe factor, and item 32 of the describe facet scale; see also, Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006; Medvedev, Siegert, Kersten, & Krägeloh, 2017) (CFI=.956, TLI=.946, RMSEA=.042) fit the data better than a similar five-factor, hierarchical model (CFI=.925, TLI=.913, RMSEA=0.046). The four-factor, hierarchical FFMQ also demonstrated evidence of strong measurement invariance across gender, grade level, and assessment wave. Finally, the FFMQ total scale score (excluding the observe facet and item 32) was correlated at each assessment wave with different social-emotional outcomes including self-compassion (rs=.64 to .68) and perceived stress scale (rs=-.64 to -.54, ps<.001).

Discussion: This study is the first to examine the psychometric properties of the FFMQ for use with American adolescents. Our results suggest that a 20-item short-form FFMQ demonstrated evidence of adequate reliability and concurrent validity with theoretically-related constructs, including self-compassion and perceived stress. Likewise, the instrument demonstrated evidence of a hypothesized four-factor, hierarchical structure shown in adult samples and measurement invariance across gender, grade level, and time. The current study provides initial support for the use of this short-form FFMQ instrument with high school-age adolescents.

Authors

• Hiba Abujaradeh, University of Pittsburgh
  Non-Presenting Author

• Blake Colaianne, Pennsylvania State University
  Non-Presenting Author

• Robert William Roeser, Pennsylvania State University
  Non-Presenting Author

• Brian M Galla, University of Pittsburgh
  Presenting Author