Published September 13, 2026. Last updated September 13, 2026.
Being kind to yourself after a mistake sounds simple until the moment actually arrives, and what usually arrives instead is the older, harsher voice. Psychology has a name and a measurement for the gap between those two responses, and a research base that has grown into meta-analyses covering tens of thousands of people. This page gathers what the Self-Compassion Scale actually measures, what the meta-analyses on well-being and psychopathology report, what the intervention trials found when researchers tried to teach the skill directly, and where no real prevalence figure exists at all.
Key takeaways
- A U.S. sample of 391 undergraduates at the University of Texas at Austin, data-collection year not stated in the paper, publication 2003, was used to build the Self-Compassion Scale’s factor structure (Neff, 2003).
- Self-Compassion Scale scores among undergraduates and community adults in the United States cluster close to a mean of 3.0 with a standard deviation near 0.60 on the scale’s 1-to-5 item average, samples span multiple studies, chapter published 2022 (Neff & Toth-Kiraly, 2022).
- Across 20 samples drawn from 14 studies, all using the Self-Compassion Scale, self-compassion correlated with psychopathology at r = -.54, a measure of association only, studies span the years leading up to the analysis, published 2012 (MacBeth & Gumley, 2012).
- Across 79 samples and 16,416 people, studies span the years leading up to the analysis, published 2015, self-compassion correlated with well-being at r = .47 (Zessin, Dickhauser, & Garbade, 2015).
- Across 27 randomized controlled trials of self-compassion interventions, population not stated as a single pooled figure in available text, studies span the years leading up to the analysis, published 2019, participants who completed a self-compassion intervention improved on self-compassion itself (Hedges g = 0.75), stress (g = 0.67), depression (g = 0.66), anxiety (g = 0.57), and self-criticism (g = 0.56), each measured against a control group (Ferrari et al., 2019).
- Across 71 journal articles and dissertations yielding 88 effect-size estimates, population not stated as a single pooled figure in available text, studies span the years leading up to the analysis, published 2015, men scored slightly higher than women on self-compassion, d = 0.18, a small gap that widened in samples with more ethnic-minority participants (Yarnell et al., 2015).
- No representative survey located for this page reports what percentage of the population “struggles with self-compassion” or scores below any clinical cutoff, because the Self-Compassion Scale’s own documentation states plainly that no such cutoff exists (Neff, Self-Compassion Scale information).
- Among 3,112 university students surveyed no later than 2021, published 2023, the 624 students classified as adaptive perfectionists reported significantly greater self-compassion than the 1,110 classified as maladaptive perfectionists (Kawamoto et al., 2023).
Cite this page
APA: Eduard, S. (2026, September 13). Self-compassion statistics: What the scale, the meta-analyses, and the trials show. StumbleTowardTheLight. https://stumbletowardthelight.com/self-compassion-statistics/
MLA: Eduard, Sebastian. “Self-Compassion Statistics: What the Scale, the Meta-Analyses, and the Trials Show.” StumbleTowardTheLight, 13 Sept. 2026, stumbletowardthelight.com/self-compassion-statistics/.
The source table
| # | Statistic | Data collected | Source type | Source |
|---|---|---|---|---|
| 1 | Self-Compassion Scale built and factor-tested on 391 University of Texas at Austin undergraduates | Collection year not stated in the paper; published 2003 | primary study | Neff, 2003, Self and Identity |
| 2 | SCS mean scores cluster near 3.0 (SD near 0.60) on the 1-to-5 item average, for undergraduates and community adults in the U.S. | Samples span multiple studies; chapter published 2022 | primary study (measure handbook chapter) | Neff & Toth-Kiraly, 2022, Handbook of Assessment in Mindfulness Research |
| 3 | The scale’s own documentation states there is no clinical cutoff for “low” or “high” self-compassion | Not a dated study; measure documentation, checked directly 2026 | primary study (measure documentation) | Neff, Self-Compassion Scale information |
| 4 | Self-compassion correlated with psychopathology at r = -.54 (95% CI -.57 to -.51); 20 samples, 14 studies, all using the SCS | Studies span the years leading up to the analysis; published 2012 | primary study (meta-analysis) | MacBeth & Gumley, 2012, Clinical Psychology Review |
| 5 | Self-compassion correlated with well-being at r = .47; 79 samples, 16,416 people | Studies span the years leading up to the analysis; published 2015 | primary study (meta-analysis) | Zessin, Dickhauser, & Garbade, 2015, Applied Psychology: Health and Well-Being |
| 6 | Self-compassion interventions raised self-compassion itself, Hedges g = 0.75, across 27 randomized controlled trials | Studies span the years leading up to the analysis; published 2019 | primary study (meta-analysis) | Ferrari et al., 2019, Mindfulness |
| 7 | Same trials reduced stress (g = 0.67), depression (g = 0.66), anxiety (g = 0.57), and self-criticism (g = 0.56) against control groups | Same as row 6 | primary study (meta-analysis) | Ferrari et al., 2019, Mindfulness |
| 8 | The same trials found large gains on eating-behavior measures (g = 1.76) and rumination (g = 1.37) | Same as row 6 | primary study (meta-analysis) | Ferrari et al., 2019, Mindfulness |
| 9 | Men scored slightly higher than women on self-compassion, d = 0.18, gap wider in samples with more ethnic-minority participants; 71 studies and dissertations, 88 effect-size estimates | Studies span the years leading up to the analysis; published 2015 | primary study (meta-analysis) | Yarnell et al., 2015, Self and Identity |
| 10 | Of 3,112 university students, adaptive perfectionists (624, 20.1%) reported significantly greater self-compassion than maladaptive perfectionists (1,110, 35.7%) | Unstated in the paper; newest cited sources 2020, published online 2022, so no later than 2021 | primary study | Kawamoto et al., 2023, The Counseling Psychologist |
| 11 | No nationally representative survey located reports a percentage of the population that “struggles with” or falls below a cutoff for self-compassion | 2026 (this page’s check) | derived (absence disclosed, per rows 2 and 3) | see rows 2 and 3 |
Self-compassion statistics: how common is low self-compassion?
No study answers that question with a population percentage. Self-compassion is measured as a continuous score on a questionnaire, producing an average and a spread for whoever answers it. The scale’s own scoring guidance states directly that there are no clinical norms or cutoffs marking someone as low or high in self-compassion (Neff, Self-Compassion Scale information). What exists instead is a group average: undergraduate and community adult samples in the United States cluster close to 3.0 on the scale’s 1-to-5 item average, with a standard deviation near 0.60, placing most people in a moderate middle band on the distribution.
How many people struggle with self-compassion?
There is no representative survey that produces this figure, and that gap is itself the honest finding. The closest real evidence is indirect: across 20 samples and 14 studies using the Self-Compassion Scale, lower self-compassion tracked with higher psychopathology at a strong correlation of r = -.54 (MacBeth & Gumley, 2012). Among 3,112 university students, the 1,110 classified as maladaptive perfectionists scored lower on self-compassion than the 624 classified as adaptive perfectionists, a comparison between two groups inside one sample (Kawamoto et al., 2023). Anyone recognizing that gap in their own reactions will find the mechanism described in why some people are so hard on themselves, and what loosens its grip in overcoming self-criticism.
Why is self-compassion so hard?
Because the people whose research profile shows the least of it are also the ones under the most internal pressure to perform. Among 3,112 university students, those classified as maladaptive perfectionists, driven by harsh personal standards paired with a strong fear of falling short, reported significantly lower self-compassion than adaptive perfectionists in the same sample (Kawamoto et al., 2023). Across 71 studies and dissertations, gender added a second layer: men scored slightly higher than women on self-compassion overall, a small gap (d = 0.18) that widened further in samples with a higher proportion of ethnic-minority participants, evidence the researchers read as a signal that social norms around self-criticism and self-care are not distributed evenly (Yarnell et al., 2015). Neither finding explains the mechanism fully; both point toward the same pattern, that self-compassion is harder to access exactly where self-criticism already runs strongest. The territory of that inner voice is covered in what the inner critic actually is.
What is the average self-compassion scale score?
Close to 3.0 out of 5. Neff and Toth-Kiraly’s handbook chapter on the Self-Compassion Scale reports that mean scores among undergraduates and community adults in the United States sit near 3.0 on the 1-to-5 item average, with a standard deviation close to 0.60, samples span multiple studies, chapter published 2022 (Neff & Toth-Kiraly, 2022), and the scale’s own scoring documentation states outright that no version of the scale carries a clinical cutoff separating “low” from “high” (Neff, Self-Compassion Scale information). The scale itself was built and factor-tested on 391 undergraduates at the University of Texas at Austin, the paper’s data-collection year unstated, publication 2003 (Neff, 2003). A score near the middle of the range is the statistical norm for both undergraduates and community adults.
Does self-compassion actually work?
The intervention evidence says yes, measured against real control groups. Across 27 randomized controlled trials, people assigned to a self-compassion intervention improved on self-compassion itself (Hedges g = 0.75), and also on stress (g = 0.67), depression (g = 0.66), anxiety (g = 0.57), and self-criticism (g = 0.56), each a moderate-to-large effect against people who received no such intervention (Ferrari et al., 2019). The same set of trials found even larger shifts on eating-behavior measures (g = 1.76) and rumination (g = 1.37). Separately, across 79 samples and 16,416 people, naturally occurring self-compassion correlated with overall well-being at r = .47, a moderate relationship measured without any intervention at all (Zessin, Dickhauser, & Garbade, 2015). Two different kinds of evidence, correlational and experimental, point the same direction toward the same conclusion. What the skill looks like in practice is covered in what mindful self-compassion involves.
What does the Self-Compassion Scale actually measure?
Three paired components: kindness toward yourself set against self-judgment, a sense of shared human experience set against isolation, and mindful awareness of a difficult feeling set against over-identifying with it (Neff, 2003). The scale asks nothing about self-esteem, achievement, or comparison to other people, the ground self-compassion versus self-esteem covers directly. The correlational figures above (r = -.54 with psychopathology, r = .47 with well-being) describe association. The randomized trials supply the stronger causal evidence, because they compare an intervention group against a control group that received none.
Methodology
Statistics on this page start from the sources named in data/citation-ledger.md for this cluster (Neff, 2003; Kawamoto et al., 2023), then add primary meta-analyses and the scale developer’s own scoring documentation, checked directly against the publishing journal or the source organization’s page. Every correlation, effect size, and sample count is read directly from the study’s own reported abstract or results section. Where a paper’s underlying fielding years were not stated in the retrievable text, this page discloses that plainly. Sources last verified September 13, 2026.
Cite this page
APA: Eduard, S. (2026, September 13). Self-compassion statistics: What the scale, the meta-analyses, and the trials show. StumbleTowardTheLight. https://stumbletowardthelight.com/self-compassion-statistics/
MLA: Eduard, Sebastian. “Self-Compassion Statistics: What the Scale, the Meta-Analyses, and the Trials Show.” StumbleTowardTheLight, 13 Sept. 2026, stumbletowardthelight.com/self-compassion-statistics/.
Frequently asked questions
What percentage of people have low self-compassion?
No representative survey reports that figure, because the Self-Compassion Scale’s own documentation states there is no clinical cutoff separating low from high self-compassion (Neff, Self-Compassion Scale information). Undergraduate and community adult samples in the U.S. average close to 3.0 out of 5, samples span multiple studies, chapter published 2022 (Neff & Toth-Kiraly, 2022).
Is self-compassion the same as self-esteem?
No. The Self-Compassion Scale measures self-kindness, common humanity, and mindfulness, none of which involve comparing yourself to others or evaluating your achievements (Neff, 2003). The full comparison lives on self-compassion versus self-esteem.
Do men or women score higher on self-compassion?
Men scored slightly higher than women on average, a small effect (d = 0.18) drawn from 71 studies and dissertations yielding 88 estimates, with the gap widening in samples with more ethnic-minority participants (Yarnell et al., 2015).
Can you actually increase your self-compassion?
Randomized controlled trials say yes. Across 27 such trials, participants raised their self-compassion scores by a large margin (Hedges g = 0.75) and improved on stress, depression, anxiety, and self-criticism compared with control groups (Ferrari et al., 2019).
How strong is the link between self-compassion and mental health?
Strong and consistent across two different kinds of evidence: a correlation of r = -.54 with psychopathology across 20 samples (MacBeth & Gumley, 2012) and a correlation of r = .47 with well-being across 79 samples and 16,416 people (Zessin, Dickhauser, & Garbade, 2015).
Related reading
References
Ferrari, M., Hunt, C., Harrysunker, A., Abbott, M. J., Beath, A. P., & Einstein, D. A. (2019). Self-compassion interventions and psychosocial outcomes: A meta-analysis of RCTs. Mindfulness, 10(8), 1455-1473.
Kawamoto, A., Sheth, R., Yang, M., Demps, L., & Sevig, T. (2023). The role of self-compassion among adaptive and maladaptive perfectionists in university students. The Counseling Psychologist, 51(1), 113-144.
MacBeth, A., & Gumley, A. (2012). Exploring compassion: A meta-analysis of the association between self-compassion and psychopathology. Clinical Psychology Review, 32(6), 545-552.
Neff, K. D. (2003). The development and validation of a scale to measure self-compassion. Self and Identity, 2(3), 223-250.
Neff, K. D. (n.d.). Self-Compassion Scale (SCS) information and scoring. Self-Compassion.org.
Neff, K. D., & Toth-Kiraly, I. (2022). Self-Compassion Scale (SCS). In O. N. Medvedev et al. (Eds.), Handbook of Assessment in Mindfulness Research. Springer.
Yarnell, L. M., Stafford, R. E., Neff, K. D., Reilly, E. D., Knox, M. C., & Mullarkey, M. (2015). Meta-analysis of gender differences in self-compassion. Self and Identity, 14(5), 499-520.
Zessin, U., Dickhauser, O., & Garbade, S. (2015). The relationship between self-compassion and well-being: A meta-analysis. Applied Psychology: Health and Well-Being, 7(3), 340-364.