In plain language
Experiential avoidance — the tendency to avoid unpleasant thoughts and feelings — is a key target of many psychological therapies, so measuring it well matters. The best multidimensional measure, the Multidimensional Experiential Avoidance Questionnaire (MEAQ), has 62 items, which is too long for many clinical and research settings. The existing 15-item short version solves the length problem but collapses the measure into a single dimension, losing information about the six distinct ways people avoid their inner experiences.
This study used a genetic algorithm — a machine-learning technique that mimics Darwinian evolution to search efficiently through an astronomically large space of possible short forms — to cut the MEAQ in half while keeping all six of its dimensions: behavioral avoidance, distress aversion, procrastination, distraction and suppression, repression/denial, and distress endurance. The authors applied the method (implemented in their free R package, GAabbreviate) to data from 7,884 Americans in a nationally representative sample.
The resulting 30-item MEAQ-30, with five items per subscale, performed virtually identically to the full 62-item version: it showed the same factor structure, the same correlations among subscales, and the same associations with mental distress, well-being, and personal strivings. It even showed similar score distributions across American census regions. The MEAQ-30 lets researchers and clinicians measure all six facets of experiential avoidance at half the participant burden, and the study showcases how machine learning can improve psychological measurement.
Key findings
- A genetic algorithm successfully shortened the 62-item Multidimensional Experiential Avoidance Questionnaire to a 30-item version (MEAQ-30) with five items for each of its six subscales.
- The abbreviation was developed and validated in a large nationally representative American sample (N = 7,884; 52% female; mean age 47.9 years).
- The MEAQ-30 performed virtually identically to the original in inter-subscale correlations, factor structure, and factor correlations.
- Zero-order correlations and unique latent associations of the six subscales with measures of mental distress, well-being, and personal strivings were nearly the same for the MEAQ-30 as for the full 62-item measure.
- Both versions showed similar distributions of subscale means across American census regions.
- The study demonstrates the utility of machine-learning methods in psychometrics, using the freely available R package GAabbreviate.
How to cite
APA
Sahdra, B. K., Ciarrochi, J., Parker, P., & Scrucca, L. (2016). Using genetic algorithms in a large nationally representative American sample to abbreviate the Multidimensional Experiential Avoidance Questionnaire. Frontiers in Psychology, 7, 189. https://doi.org/10.3389/fpsyg.2016.00189
BibTeX
@article{sahdra2016using,
author = {Sahdra, Baljinder K. and Ciarrochi, Joseph and Parker, Philip and Scrucca, Luca},
title = {Using Genetic Algorithms in a Large Nationally Representative American Sample to Abbreviate the Multidimensional Experiential Avoidance Questionnaire},
journal = {Frontiers in Psychology},
year = {2016},
volume = {7},
pages = {189},
doi = {10.3389/fpsyg.2016.00189}
}
Related work
- All publications by Joseph Ciarrochi (searchable, with free PDFs)
- Process-Based Therapy & Idionomic Analysis
Author: Joseph Ciarrochi (ORCID 0000-0003-0471-8100). Free copy hosted with permission for scholarly use. Please cite the published version via the DOI above.