The Self-Compassion Scale – Short Form (SCS-SF) is a 12-item self-report measure of self-compassion designed for adults aged 18 years and over (Raes et al., 2011). It was developed as a brief alternative to Neff’s (2003) original 26-item SCS.
Self-compassion is a way of relating to oneself with kindness, a sense of shared human experience, and balanced awareness during moments of suffering, perceived inadequacy, or failure (Neff, 2003). It is considered the same compassion one would naturally extend to a struggling friend, turned inward. This stands in contrast to three less adaptive responses to suffering (Neff, 2003): harsh self-criticism, a sense of isolation in one’s suffering, and over-identification with painful thoughts. Self-compassion is distinct from self-pity, self-indulgence, and self-esteem. Furthermore, it does not depend on positive self-evaluation, comparison with others, or excusing harmful behaviour, but rather on meeting one’s own suffering with the same care one would offer another (Neff, 2003). Higher self-compassion is consistently associated with lower depression, anxiety, and rumination, and with greater wellbeing, optimism, and resilience (Hayes et al., 2016; Adu et al., 2024).
In addition to producing a Total Score, the SCS-SF includes two subscales (Hayes et al., 2016):
Example SCS-SF Item
Low self-compassion is common across many clinical presentations, including depression, anxiety, trauma, eating disorders, chronic illness, and perfectionism (MacBeth & Gumley, 2012; Neff, 2023). The SCS-SF is suitable for use in individual therapy, group programs, research, and fits alongside most therapy approaches rather than being tied to a particular one. It is not a diagnostic tool and does not indicate whether a disorder is present; its purpose is to provide a quick, trackable read on self-compassion as a treatment target.
Administered at intake, the SCS-SF establishes the baseline level of self-compassion and indicates whether it is a relevant focus for treatment. Repeated across treatment, it tracks whether levels of self-compassion are shifting. The two subscales help direct focus: individuals high on Self Disparagement may benefit from work that softens self-attack, while clients low on Self Care may benefit from skill-building around self-kindness and balanced awareness. The score also provides a concrete way to talk with individuals about self-compassion – what it is, why it matters, and how they relate to themselves under stress.
Self-compassion is the central focus of compassion-focused therapy (CFT) and mindful self-compassion (MSC) training, but the SCS-SF is compatible with most therapy approaches, including mindfulness-based cognitive therapy (MBCT), acceptance and commitment therapy (ACT), schema therapy, and any approach where self-criticism, shame, or harshness toward the self are part of clinical consideration. Across these approaches, the SCS-SF supports making self-compassion an explicit and trackable focus of treatment.
The SCS-SF total average score is an overall indicator of self-compassion. It is the average of all 12 items (range 1 to 5), made up of the six Self Care items and the six Self Disparagement items. The Self Disparagement items are reverse-scored first so that higher totals reflect more self-compassion. As a result, a high total reflects high Self Care together with low Self Disparagement.
Subscale interpretation is provided only to support clinical conversation about the relative balance between Self Disparagement and Self Care dimensions of self-compassion. Both subscales are total averages using the same scale as the total score (1-5).
Self Disparagement (items 1, 4, 8, 9, 11, 12): captures harsh self-criticism, isolation in failure, and fixation on personal inadequacy. Unlike the total score, in which the Self Disparagement items are reverse scored, higher Self Disparagement scores on their own indicate more harsh, isolating self-responses (e.g., thoughts like “there’s something wrong with me”, withdrawal when distressed). Lower scores reflect fewer self-attacking responses, but higher total Self Compassion score.
Self Care (items 2, 3, 5, 6, 7, 10): captures self-kindness, recognition of common humanity, and a balanced mindful stance. Higher Self Care scores indicate more warm, balanced self-responses. Respondents that score high on Self Care find thoughts such as “this is hard, and that’s understandable” relatively accessible, and may seek comfort, support, or rest in response to distress. Lower may indicate individuals push through distress without acknowledgement or make limited use of self-soothing strategies.
A percentile is presented for the average total score and for each subscale, indicating how the respondent’s score compares with the community reference sample (Adu et al., 2024, New Zealand subsample). A community percentile of 50 represents a typical score within the community reference sample. Importantly, the subscales run in opposite directions, so a high percentile does not mean the same for each average score:
The results table displays the average score (1 to 5), and community percentile for the Total Score and subscales scores.
On the first administration, two bar charts are shown. The first graph plots the respondent’s total average score against the community sample, so clinicians can quickly see where they sit relative to most adults. The second graph compares the two subscales, Self Disparagement and Self Care, with colour shading on each bar indicating how the scores sit relative to the community sample.
On repeat administrations, a multi-administration line plot replaces the single-administration graphs for the total score and subscales, plotting the average scores over time, so that change can be seen across administrations.
The SCS-SF was constructed by Raes and colleagues (2011) by selecting two items from each of the six subscales of the original 26-item Self-Compassion Scale (Neff, 2003). Item selection was guided by high correlations with both the long-form total and the relevant subscale, while preserving content breadth. Construction and cross-validation were conducted in two Dutch-speaking samples (n = 271 first-year psychology students; n = 185 community adults recruited online), and the resulting 12-item scale was then validated in a third, English-speaking sample of 415 US university students. The total score on the short form correlated near-perfectly with the long form (r ≥ 0.97 across samples), supporting its use as a brief alternative for measuring overall self-compassion.
The original six subscales score poorly when derived from the 12 SCS-SF items (α = .54–.75) and are not recommended (Raes et al., 2011). The two SCS-SF subscales (Self Care and Self Disparagement) have acceptable reliability (α = .79–.87, Hayes et al., 2016; .72–.85, Babenko & Guo, 2019) and reflect a clinically meaningful split between positive and negative self-responding, but do not have strong psychometric support as standalone scales.
The SCS-SF total is the primary output. It has good internal consistency (α = .86, n = 415, Raes et al., 2011; α = .85, Hayes et al., 2016; α = .86, Babenko & Guo, 2019), correlates near-perfectly with the long-form SCS (r = .97; Raes et al., 2011), and fits a unidimensional Rasch model (PSI = .71; Adu et al., 2024). Short-interval test-retest reliability is good in translated versions (ICC = .89 over 1–2 weeks in Spanish, n = 103, Garcia-Campayo et al., 2014; ICC = .84 over 2 weeks in Swedish, n = 121, Alfonsson et al., 2023), though comparable data have not been published for the English-language SCS-SF.
Convergent validity has been demonstrated through expected correlations between the SCS-SF and a range of clinically relevant constructs. Adu et al. (2024) reported a strong positive correlation with optimism (r = .51) and moderate correlations with positive affect (r = .37), psychological distress (r = −.43), and negative affect (r = −.39). In the Hayes et al. (2016) clinical sample, the SCS-SF total score showed strong inverse correlations with depression (r = −.67), social anxiety (r = −.57), and generalised anxiety (r = −.51); moderate inverse correlations with hostility (r = −.45), academic distress (r = −.41), and eating concerns (r = −.33); and a small inverse correlation with family concerns (r = −.27). In Babenko and Guo (2019), the SCS-SF correlated moderately with student exhaustion (r = −.44) and weakly with student engagement (r = .24).
Known-groups validity is supported in clinical samples. Hayes et al. (2016) demonstrated that college counselling clients with prior suicidal ideation, prior suicide attempt, or prior non-suicidal self-injurious behaviour scored significantly lower on the SCS-SF than clients without such histories, with moderate to large effect sizes (Cohen’s d = .56 to .68). Discriminant validity is supported in the same clinical sample by the absence of correlation between the SCS-SF and non-problematic perfectionism (high personal Standards; r = .04; Hayes et al., 2016), exactly as predicted.
The reference sample used to anchor percentile interpretation in NovoPsych is a New Zealand subsample of community adults from Adu et al. (2024). Although Adu et al. (2024) demonstrated measurement invariance across four countries (Ghana, Germany, India, and New Zealand), NovoPsych selected New Zealand because English is its primary language.
The NZ subsample descriptives used to anchor NovoPsych’s community percentiles are not reported in the published paper. P. Adu (personal communication, 18 May 2026) provided the NZ sample size (n = 413) and the Self Care and Self Disparagement subscale means and standard deviations. The Total mean and standard deviation were calculated from these subscale statistics, with the Total standard deviation derived using the correlation between Self Care and Self Disparagement reported by Babenko and Guo (2019; r = −.59). NovoPsych converts raw scores to percentiles against these NZ community values.
Descriptive statistics for the New Zealand community sample are shown below, first as raw scores and then as average scores (1 to 5):
Support for SCS-SF ability to detect change comes from Smeets et al. (2014), who used the 12-item SCS-SF (Dutch version) to evaluate a 3-week self-compassion intervention with female college students and found significantly greater increases in self-compassion in the intervention group compared with an active time-management control. No formal reliable change index has been established for the English-language SCS-SF. Adu et al. (2024) note that the SCS-SF’s Person Separation Index of .71 is adequate for group-level measurement but not sufficient for within-group assessment of individual change. NovoPsych therefore interprets total score changes against the distribution-based MID guidance and recommends interpreting individual change estimates with these caveats in mind.
When SCS-SF scores are available from multiple administrations, changes in the total score can indicate the effectiveness of treatment or shifts in self-compassion over time. In NovoPsych, change is interpreted against a Minimum Important Difference (MID; Norman et al., 2003) of 0.4 on the 1–5 average scale, equivalent to a 5-point change on the raw 12–60 total score (half the community SD, rounded up to the nearest whole total-score point). Changes are presented as change in average score and are categorised as:
Self-compassion is the capacity to relate to oneself with warmth, understanding, and balance in times of distress, regardless of how one is performing. While self-esteem depends on positive self-evaluation and comparison with others, self-compassion does not and remains accessible in the face of failure or struggle.
Clinicians often find it helpful to discuss what higher and lower scores suggest, and to explore which aspects of self-compassion across the two subscales feel most and least accessible to the client. Results are best framed as a current snapshot rather than a fixed trait, since self-compassion can shift with practice.
A low score does not indicate a fixed problem; it often reflects a current relationship with the self that can shift with focused work. Very low scores tend to be common in clients presenting with depression, anxiety, perfectionism, or chronic self-criticism. Self-compassion typically responds well to brief, structured interventions (e.g., CFT, MSC, brief writing tasks). The subscale pattern can complement the total at the low end, since a low Self Care score and a high Self Disparagement score point to different therapeutic emphases.
Symptom measures like the DASS-21, PHQ-9, or GAD-7 tell you how distressed a client is, while the SCS-SF tells you how they tend to relate to themselves when they are struggling. Used together, they give a fuller picture than either does alone, and this pairing is often most useful when symptoms are not shifting as expected. A client can be engaging fully in therapy yet doing so in a harsh, self-critical way, and a low self-compassion score can flag this when a symptom measure alone would not, pointing to a clear treatment target and a possible compassion-focused approach.
The SCS-SF can also be tracked over time alongside symptom measures. If distress is easing and self-compassion is building, that supports the current direction; if symptoms stay stuck and self-compassion remains low, that is a useful prompt to revisit the formulation.
The SCS-SF subscale profile is a useful guide to where to focus. An elevated Self Disparagement score suggests the client may benefit more from work that softens self-attack, such as chair work, decentring from the self-critic, and CFT threat-system formulations. A low Self Care score suggests the client may benefit more from skill-building around warmth and balanced awareness, such as the self-compassion break, soothing-rhythm breathing, and loving-kindness practice. Repeat administration across treatment lets the clinician and client see whether these practices are shifting self-compassion as intended.
For some clients, particularly those with histories of trauma, shame, attachment difficulties, or chronic self-criticism, opening to self-compassion can initially feel destabilising rather than soothing. This experience is sometimes called “backdraft”: allowing kindness toward the self can release previously suppressed emotional pain, much like opening a door to a smouldering room causes a sudden inrush of flame. A related phenomenon is described as “fear of compassion.” A very low SCS-SF score, or a paradoxical drop in self-compassion early in treatment, may signal backdraft rather than failure of the intervention.
Practical responses include pacing self-compassion practices more slowly, using a “compassionate other” entry point (offering compassion to a friend before turning it inward), validating that the distress is a normal part of opening to suffering rather than a sign that something is wrong, and pairing self-compassion work with grounding or distress-tolerance skills.
Raes, F., Pommier, E., Neff, K. D., & Van Gucht, D. (2011). Construction and factorial validation of a short form of the Self-Compassion Scale. Clinical Psychology and Psychotherapy, 18(3), 250–255. https://doi.org/10.1002/cpp.702
Adu, P., Popoola, T., Bartholomew, E., Iqbal, N., Roemer, A., Jurcik, T., Collings, S., Aspin, C., Medvedev, O. N., & Simpson, C. R. (2024). Enhancing the precision of the Self-Compassion Scale Short Form (SCS-SF) with Rasch methodology. Mindfulness, 15, 2779–2793. https://doi.org/10.1007/s12671-024-02462-y
Alfonsson, S., Wolf-Arehult, M., Winai, E., Collin, E., & Isaksson, M. (2023). The Self-Compassion Scale–Short Form: Psychometric evaluation in one non-clinical and two clinical Swedish samples. Clinical Psychology & Psychotherapy, 30, 631–642. https://doi.org/10.1002/cpp.2830
Babenko, O., & Guo, Q. (2019). Measuring self-compassion in medical students: Factorial validation of the Self-Compassion Scale–Short Form (SCS-SF). Academic Psychiatry, 43, 590–594. https://doi.org/10.1007/s40596-019-01095-x
Garcia-Campayo, J., Navarro-Gil, M., Andrés, E., Montero-Marin, J., López-Artal, L., & Demarzo, M. M. P. (2014). Validation of the Spanish versions of the long (26 items) and short (12 items) forms of the Self-Compassion Scale (SCS). Health and Quality of Life Outcomes, 12, 4. https://doi.org/10.1186/1477-7525-12-4
Hayes, J. A., Lockard, A. J., Janis, R. A., & Locke, B. D. (2016). Construct validity of the Self-Compassion Scale–Short Form among psychotherapy clients. Counselling Psychology Quarterly, 29(4), 405–422. https://doi.org/10.1080/09515070.2016.1138397
López, A., Sanderman, R., Smink, A., Zhang, Y., van Sonderen, E., Ranchor, A. V., & Schroevers, M. J. (2015). A reconsideration of the Self-Compassion Scale’s total score: Self-compassion versus self-criticism. PLoS ONE, 10(7), e0132940. https://doi.org/10.1371/journal.pone.0132940
Muris, P., & Petrocchi, N. (2017). Protection or vulnerability? A meta-analysis of the relations between the positive and negative components of self-compassion and psychopathology. Clinical Psychology & Psychotherapy, 24(2), 373–383. https://doi.org/10.1002/cpp.2005
Neff, K. D. (2003). The development and validation of a scale to measure self-compassion. Self and Identity, 2(3), 223–250. https://doi.org/10.1080/15298860309027
Neff, K. D., Long, P., Knox, M. C., Davidson, O., Kuchar, A., Costigan, A., Williamson, Z., Rohleder, N., Tóth-Király, I., & Breines, J. G. (2018). The forest and the trees: Examining the association of self-compassion and its positive and negative components with psychological functioning. Self and Identity, 17(6), 627–645. https://doi.org/10.1080/15298868.2018.1436587
Neff, K. D. (n.d.). Self-compassion. https://self-compassion.org/
Neff, K. D. (2023). Self-Compassion: Theory, Method, Research, and Intervention. Annual Review of Psychology, 74, 193–218. https://doi.org/10.1146/annurev-psych-032420-031047
Norman, G. R., Sloan, J. A., & Wyrwich, K. W. (2003). Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation. Medical Care, 41(5), 582–592. https://doi.org/10.1097/01.MLR.0000062554.74615.4C
Raes, F., Pommier, E., Neff, K. D., & Van Gucht, D. (2011). Construction and factorial validation of a short form of the Self-Compassion Scale. Clinical Psychology and Psychotherapy, 18(3), 250–255. https://doi.org/10.1002/cpp.702
Smeets, E., Neff, K., Alberts, H., & Peters, M. (2014). Meeting suffering with kindness: Effects of a brief self-compassion intervention for female college students. Journal of Clinical Psychology, 70(9), 794–807. https://doi.org/10.1002/jclp.22076
