PERMA Holds Up Across 31,966 People (Butler & Kern, 2016)

The Question

When Seligman published his five-pillar model of flourishing in 2011, he didn't have a way to measure it. Could a single short test capture all five separately?

What They Found

• The five-factor structure (Positive emotion, Engagement, Relationships, Meaning, Accomplishment) replicated in every one of the 7 samples tested
• Internal reliability was strong: Cronbach's alpha was above 0.80 on every pillar
• Pillars correlated with each other only moderately (around r = 0.5), meaning each one really does capture a distinct dimension of flourishing
• The 23-item version (15 pillar items plus 8 supplementary measures of health, negative emotion, loneliness, and overall happiness) proved short enough for individual coaching and detailed enough for research
• The international mean across the 31,966-person sample landed at 6.98 out of 10, with most people noticeably stronger on some pillars than others

How They Tested It

• Generated hundreds of candidate items, then refined them across 7 successive samples before narrowing to the final 23
• Final pooled validation on 31,966 people, primarily from English-speaking countries
• Ran confirmatory factor analysis to test the 5-pillar structure against alternative models (single factor, three factors, etc.)
• Compared the PERMA-Profiler against established well-being scales to verify it correlated with what it should and diverged from what it shouldn't

Caveats

• Validation skewed toward Western, English-speaking populations. Adaptations for non-Western languages and cultures came later
• Pillars correlate enough that the overall PERMA score is hard to distinguish from a generic well-being rating (a critique formalized in Goodman et al., 2018)
• The supplementary scales (health, negative emotion, loneliness, happiness) are reported alongside but are not part of the PERMA model itself