See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/384810502

Translation and Validation Study of the Korean Self-Compassion Scale

Article in Mindfulness · October 2024

DOI: 10.1007/s12671-024-02453-z

CITATIONS READS

0 33

4 authors, including:

Jeong Eun Cheon Janet D Latner

Yonsei University University of Hawaiʻi at Mānoa
12 PUBLICATIONS 62 CITATIONS 175 PUBLICATIONS 9,748 CITATIONS

SEE PROFILE SEE PROFILE

Young-Hoon Kim
Hanyang University
751 PUBLICATIONS 23,672 CITATIONS

SEE PROFILE

All content following this page was uploaded by Janet D Latner on 12 October 2024.

The user has requested enhancement of the downloaded file.

Mindfulness
https://doi.org/10.1007/s12671-024-02453-z

ORIGINAL PAPER

Translation and Validation Study of the Korean Self‑Compassion Scale

Si Woo Chae1 · Jeong Eun Cheon2 · Janet D. Latner1 · Young‑Hoon Kim2

Accepted: 17 September 2024

© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024

Abstract
Objective Several studies have been conducted to examine the factor structure of the Self-Compassion Scale in multiple
languages. A preliminary review of the existing Korean version of the Self-Compassion Scale (SCS) indicated that some
items were less accurately translated, necessitating a retranslation. The goal of the current study was to provide a reliable
version of the K-SCS that can be used in further research on self-compassion, providing an accurate means of assessing the
nature and correlates of self-compassion in a Korean population.
Method We translated K-SCS to address conceptual and methodological concerns using approved translation procedures.
Upon translating the SCS, a total of 510 Korean participants completed the survey via online survey platform. We explored
the factor structure of the scale using confirmatory factor analysis (CFA) and exploratory structural equation modeling
(ESEM). We also investigated discriminant and convergent validity of the K-SCS.
Results Findings from the current study supported the use of 6-factor correlated and bifactor models using an ESEM
approach, indicated by excellent goodness of fit indices. The K-SCS was negatively related to measures of negative emotions
and positively related to measures of life satisfaction and happiness.
Conclusions The current study provides a reliable K-SCS that can be used to assess the 6 subscales as well as the total score
of the SCS. Findings support the use of ESEM approach when examining the factor structure of the SCS, which has been
increasingly used in validating translated versions of the SCS.
Preregistration This study is not preregistered.

Keywords Self-compassion · Self-compassion scale · South Korea · Confirmatory factor analysis · Exploratory structural
equation modeling

Self-compassion, an Eastern philosophical concept derived Mindfulness is understood as recognizing one’s current
from Buddhist philosophy, is an attitude focused on showing feelings, thoughts, and situations without denying or over-
compassion towards the self (Neff, 2003a, b). According to identifying with those painful experiences; over-identifica-
Neff (2003a, b), self-compassion has three components: (a) tion is an opposite construct that describes an exaggerated
self-kindness, (b) mindfulness, and (c) common humanity, emotional response when exposed to painful experiences
with each component characterized by positive and nega- (Neff, 2003a, b). Common humanity is defined as seeing
tive poles. Self-kindness refers to providing understanding, life experiences (including failures, flaws, and painful situ-
support, and warmth to the self in the midst of failures and ations) as shared with all human beings. In contrast to com-
difficult times (Neff, 2003a, b). The counterpart of self-kind- mon humanity, isolation is defined as feeling different and
ness is self-judgment, which involves harsh self-criticism disconnected from others in the midst of difficulties (Neff,
when faced with difficulties and perceived inadequacies. 2003a, b).
Neff (2003a) conceptualizes self-compassion as showing
understanding and care for oneself when faced with suffer-
* Si Woo Chae ing and difficulties to generate a sense of well-being that is
[email protected]
rooted in the experience of being human. Self-compassion
1
Department of Psychology, University of Hawai’i at Mānoa, was negatively associated with self-criticism, depression,
2530 Dole St. Sakamaki C 400, Honolulu, HI 96822, USA anxiety, rumination, thought suppression, and neurotic per-
2
Department of Psychology, Yonsei University, Seoul, fectionism, and positively correlated with life satisfaction
South Korea

Vol.:(0123456789)
 Mindfulness

and social connectedness (Neff, 2003b). Self-compassion (Petrocchi et al., 2014), German (Hupfeld & Ruffieux, 2011),
has also been reported to be related to optimism and hap- Portuguese (Costa et al., 2015), and Dutch (López et al.,
piness, as well as personal initiative, curiosity, and explo- 2015) versions of the SCS suggested that a single higher-
ration (Neff, 2009; Neff et al., 2007), and life satisfaction order factor may not be the optimal method of examining the
(Anggraeni and Kurniawan, 2012; Arimitsu & Hofmann, overarching construct of self-compassion. Although transla-
2015). Self-compassion is not only associated with posi- tion issues could be one possible explanation for these dif-
tive psychological variables, but also negatively associated ferent factor structures, a study by Williams et al. (2014)
with psychopathology (e.g., Diedrich et al., 2016; Gilbert using the original SCS failed to replicate the factor struc-
& Procter, 2006), suggesting self-compassion as a predic- ture found in the initial validation study by Neff (2003b). In
tive correlate of mental health. A meta-analysis conducted this study, poor goodness of fit index was observed in the
by MacBeth and Gumley (2012) concluded that self-com- hierarchical 6-factor structure, and an acceptable goodness
passion is negatively associated with common expressions of fit index was indicated for the 6-factor model, suggest-
of psychopathology such as depression, anxiety, and stress. ing that the SCS may be used to assess six facets of self-
With the burgeoning literature on self-compassion and compassion independently rather than the overall level of
its putative benefits on mental health and psychological self-compassion.
well-being, advancing our understanding of the construct Due to the concerns related to the factor structure of the
of self-compassion by exploring optimal ways to opera- SCS, several studies have attempted to explore the psycho-
tionalize and assess self-compassion is crucial. The Self- metric properties of the SCS. Several studies have suggested
Compassion Scale (SCS; Neff, 2003b) is the most widely the use of a 2-factor model in which the SCS is conceptual-
used scale developed to measure thoughts, emotions, and ized to have two dimensions: (1) self-compassion, which
behaviors related to showing compassion towards the self. is comprised of self-kindness, mindfulness, and common
The factor structure of the SCS has also received a great humanity, and (2) self-coldness, which entails self-judg-
deal of research interest. It is imperative that studies on ment, over-identification, and isolation (e.g., Costa et al.,
self-compassion with different populations explore the fac- 2015; López et al., 2015). In this context, positively phrased
tor structure of the SCS because self-compassion can be items are summed to reflect the overall level of self-compas-
a meaningful psychological construct in explaining how sion, and negatively phrased items are summed to indicate
individuals respond to negative life events. Understanding the overall level of self-coldness. This aligns with Gilbert’s
the dimensions of self-compassion may inform the focus of (2005) theory of social mentalities, which states that there
psychoeducation or self-compassion interventions and assist are two distinct processing systems involved in how people
in assessing the progress of patients. interact with the world: the threat-defense and safeness sys-
In the original study of the development of the SCS (Neff, tems. According to Gilbert (2005), the internal processes
2003b), it was initially hypothesized that the measure would involved in how individuals treat themselves vary depend-
indicate a 3-factor structure (i.e., self-kindness, mindfulness, ing on which system is being activated. Further research
and common humanity). However, the confirmatory factor by Gilbert et al. (2011) indicated that the safeness system
analysis (CFA) and exploratory factor analysis (EFA) of is associated with self-compassion while the threat-defense
the three components of self-compassion resulted in 2 fac- system is related to self-coldness. Likewise, other studies
tors each, representing a 6-factor structure (Neff, 2003b). raised concerns regarding the inclusion of items related to
Neff (2003b) theorized that this may be due to the presence uncompassionate attitude (i.e., self-judgment, isolation, and
of positive and negative aspects of each facet (e.g., self- over-identification) because these items reflect vulnerabil-
kindness vs. self-judgment). Neff (2003b) noted that it is ity and psychopathological symptoms while compassionate
not uncommon for positive and negative items to load on items (i.e., self-kindness, mindfulness, common humanity)
separate factors. In addition to the 6-factor structure, Neff reflect healthy psychological functioning (Muris & Otgaar,
(2003b) reported that the intercorrelations between the 2020; Muris et al., 2018). Proponents of the 2-factor struc-
factors can be explained by a single higher-order factor of ture of the SCS particularly raise concerns regarding the
self-compassion. Thus, the SCS has 6 subscales that can be use of a total score of the SCS. These inconsistent findings
summed to create a total score that represents the overarch- suggest ambiguity in how much each component of the SCS
ing construct of self-compassion. contributes to the total score and to other relevant external
After the introduction of the SCS, numerous studies variables (Smith et al., 2009).
were conducted to replicate its factor structure across dif- In response to the ongoing concern regarding the use of
ferent populations, with inconsistent findings. For example, SCS to represent compassionate attitude towards oneself,
Chinese (Chen et al., 2011) and Spanish (Garcia-Campayo Neff (2016b) further clarified that reduced uncompassion-
et al., 2014) versions of the SCS suggested the single higher- ate attitude is a key part of the definition of self-compas-
order factor model. On the other hand, results from Italian sion. Neff (2016b) emphasized that compassionate attitude
Mindfulness

towards oneself not only entails increased self-kindness, 2-bifactor models. Results from this study supported the use
mindfulness, and common humanity, but also includes of the bifactor as well as the 6-factor correlated model using
reductions in their opposing components (i.e., self-judg- an ESEM framework for the newly translated version of the
ment, over-identification, and isolation, respectively). Neff Korean SCS.
(2016b) underscored that the subscales for uncompassionate Several studies used the Korean version of the SCS (Kim
attitude form their own dimensions, which she explained et al., 2008) to further explore the factor structure of the SCS
as the theoretical reasons for proposing the use of 6 sub- using various suggested models (e.g., single factor, 2-factor,
scales in addition to the total score. In addition to theoretical 3-factor, 6-factor, single bifactor, etc.). For instance, Lee and
explanations, Neff (2016a) introduced the bifactor model to Lee (2010), Kim and Kim (2019), and Ku and Jung (2016)
address the growing number of inconsistent findings about examined the factor structure of the Korean version of the
the dimensions of the SCS and justify the use of overall self- SCS (Kim et al., 2008) and supported the 6-factor structure
compassion score. According to Neff (2016a), in a single of the scale, which indicates the use of 6 subscales scores.
higher-order model, self-compassion explains the intercor- On the other hand, other studies reported different findings.
relations of the 6 subscales, with an assumption that self- Shin (2018) tested eight different models in which only the
compassion is not directly associated with each item in the dual factor model was supported. In this dual factor model,
SCS (Reise et al., 2010). On the other hand, in the bifac- each item loaded onto one of the 6 factors while isolation,
tor model, self-compassion is assumed to directly impact self-judgment, and over-identification factors also converged
individual items in the SCS and how these individual items onto a latent factor (i.e., self-coldness). Another study (Park
form the subscales. Neff (2016a) reclaimed that the bifactor et al., 2020) conducted several models and supported a sec-
model is a more theoretically coherent conceptualization of ond order model, which indicates that the Korean version of
the construct because self-compassion directly impacts the the SCS is comprised of the 6 domain-specific factors and
ways individuals respond to their suffering, difficult situ- two general factors (i.e., self-compassion and self-coldness).
ations, or perceived flaws, which are represented by each Although most validation studies of the Korean ver-
item in the SCS. sion of the SCS used the translated version by Kim et al.
Neff et al. (2017) further examined the factor structure (2008), a few studies attempted to translate and examine
of the SCS using various statistical models. Based on their the factor structure of the SCS. Jo and Kim (2011) trans-
findings, they suggested that future attempts to translate and lated the SCS using back-translation method and conducted
validate the SCS should use a single-factor bifactor model principal components analyses to examine the validity of
(also known as single-bifactor model) and a 6-factor cor- the use of the scale for Korean population. Findings sup-
related model when evaluating model fit. Neff et al. (2019) ported for the 6-factor structure, with the removal of items
recently explored the factor structure of the SCS in 20 dif- that neither represented the construct of self-compassion
ferent populations across 16 countries and reported that the nor the domain specific factors. Additionally, some of the
findings further supported the use of the single-bifactor retained items loaded onto a domain specific factor differ-
model and 6-factor structure model using an exploratory ent from that of the original SCS (Neff 2003b). For exam-
structural equation modeling (ESEM) framework. The ple, an item that was originally under mindfulness loaded
authors concluded that the single-bifactor model indicates onto the self-kindness domain while some other items that
that the overall score of SCS can be used to measure the gen- were originally classified under over-identification loaded
eral construct of self-compassion while a 6-factor correlated onto the isolation domain. Another study by Jin and Lee
model suggests that the SCS can be used to examine the (2009) translated the SCS and conducted CFA, which
6 subscales of constituent components of self-compassion. supported the 6-factor structure as well as the hierarchi-
Despite concerns related to the factor structure of the cal factor structure. Most recently, Park (2023) translated
SCS, it remains one of the most widely used self-report the SCS with an emphasis on readability and accessibility
measures for compassionate attitude towards oneself. In for Korean adolescents. The findings supported the use of
the context of South Korea, Kim et al. (2008) were the 6-factor model as well as the 2 bifactor model, which indi-
first researchers to translate and examine the psychometric cates the use of 6 subscales as well as two general scales
properties of the SCS in Korean. CFA was used to examine (i.e., compassionate self-responding for self-kindness,
model fit of the 3-factor structure, 6-factor structure, and a mindfulness, and common humanity subscales; uncompas-
single higher-order structure of self-compassion. The 6-fac- sionate self-responding for self-judgement, over-identifica-
tor structure provided a moderate model fit. Another study tion, and isolation subscales).
(Neff et al., 2019) using the same data collected by Kim These studies illustrate that since the introduction of
et al. (2008) examined the factor structure of the Korean the Korean version of the SCS, several studies have been
version of the SCS using various statistical models including conducted in Korea to examine the validity of the SCS for
1-factor, 2-factor, 6-factor correlated, single-bifactor, and Korean population. Although a few studies translated their
 Mindfulness

own version of the scale (Jo & Kim, 2011; Shin, 2018), these warranted to clarify some of the original version’s connota-
studies did not explain the reasons for retranslating the SCS. tions, and to capture American idioms more accurately. It
Moreover, most studies used the Korean version of the SCS was hypothesized that less accurately translated items may
translated by Kim et al. (2008) without assessing the accu- interfere with responses to the SCS and may not measure the
racy of the translation. When conducting a validation study intended construct of self-compassion. Furthermore, when
for translated measures, the initial step should be to review assessing the factor structure of the SCS, misleading items
the accuracy of the translated items. This ensures that any may result in less accurate factor loadings or items load-
discrepancies observed when examining the factor structure ing onto different factors. This can compromise the validity
of a measure can be attributed to reasons other than misin- of the measure and lead to incorrect interpretations of the
terpretation of the items. By addressing potential translation underlying constructs.
inaccuracies early in the process, researchers can ensure that Therefore, the current study aimed to retranslate the
the data collected truly reflects the intended messages of the translation of the Korean version of the SCS and exam-
original measure, thereby reducing the possibility of less ine its factor structure. Careful examination of the exist-
accurate responses. Therefore, the current study aimed to ing version of the Korean SCS identified several problems
carefully review the Korean version of the SCS translated by with the translation and suggested that retranslation of the
Kim et al., (2008) through feedback sessions before examin- measure was warranted to clarify the connotations of cer-
ing the factor structure of the Korean version of the SCS. tain words, and more accurately capture American idioms.
As an initial step, the accuracy of the previously trans- Upon retranslation of the K-SCS, the current study aimed to
lated K-SCS (Kim et al., 2008) was reviewed to assess explore 2-factor, 6-correlated, and single-bifactor CFA and
whether retranslation was warranted. Feedback on the ESEM models to investigate the factor structure of the newly
Korean version of the SCS (Kim et al., 2008) was collected translated version of the Korean SCS. Using the most opti-
from 15 native Korean speakers and 15 Koreans who are mal model based on the results, convergent and discriminant
bilingual in English and Korean. For native Korean speakers, validity were assessed by exploring the relationship between
the eligibility criteria to participate in the feedback discus- self-compassion and mental health indices.
sion were as follows: (1) born and raised in Korea and (2)
earned B.S or B.A in a university in Korea. For bilingual
Koreans, the eligibility criteria were as follows: 1) received Method
at least part of their secondary education in the US and (2)
earned a B.S or B.A in the US. The purpose of collecting Participants
feedback from bilingual Koreans who had received part of
their secondary education in the US and graduated from Participants were recruited from an online survey portal,
university in the US was to examine whether the translated with care taken to incorporate measures that may reduce
phrases are captured better by English speakers who may be some of the shortcomings of this data source. As a part of
more familiar with phrases and idioms that are characteristic the screening procedure, attention checks were used (e.g., an
of English. item stating “For this item, please click on ‘strongly disa-
Discussion groups were held by the researcher separately gree’”). Two attention check items were included. Partici-
for these two groups, evaluating the meaning and clarity of pants who failed to pass attention check items were com-
the Korean version of the SCS. Feedback from these groups pensated for completing the survey but excluded from data
indicated several concerns, with participants noting that the analyses. The datasets collected for the current study are
wording of some sentences was confusing, vague, and/or available from the corresponding author upon reasonable
awkward. Some of the translated items were also deemed request.
to convey different meanings from the original items. For Inclusion criteria were Korean citizenship and age 18
example, the use of the Korean word “gotong (고통)” that years and older. Among 837 respondents, 209 failed to
was used to refer to “suffering” conveys a sense of physical answer the attention check items correctly, 32 did not meet
pain, contrary to the broader meaning of the term in Eng- the qualifications for the study (e.g., younger than 18 years
lish that may refer to emotional, psychological, and physical old), and 86 did not fully complete the survey. Overall, the
challenges. The phrase “down and out” was translated as final sample consisted of 510 participants (47.50% male)
“galpangjilpang (갈팡질팡)”, which was understood by both with a mean age of 43.32 (SD = 13.08).
native and bilingual Koreans as emotional ups and downs.
Furthermore, “failings” was translated as “silpae (실패)”, Translation Procedure
which means failure, rather than shortcomings or weak-
nesses. This preliminary examination of the Korean version Translators for the study were volunteers drawn from
of the SCS suggested that retranslation of the measure was alumni of Emory University or Yonsei University, who
Mindfulness

met the following criteria: the individual has (a) lived in Negative Emotions
both countries for at least 4 years during middle school,
high school, and/or college, (b) is proficient in both lan- The Depression, Anxiety, and Stress Scale-21 (DASS-21;
guages, and (c) graduated from college. Based on the con- Lovibond & Lovibond, 1995) is a 21-item version of the
cerns raised during the feedback sessions, retranslation of original 42-item scale. DASS-21 assesses three related nega-
the entire measure was conducted following the guidelines tive emotional states of depression, anxiety, and tension/
provided by Brislin (1970) and Sinaiko and Brislin (1973). stress. Items are rated on a 5-point scale (0 = did not apply
Additionally, equivalence problems that frequently arise in to me at all and 5 = applied to me very much or most of the
the process of translation were considered throughout the time). Subscale scores are obtained by adding correspond-
process of translating SCS (Sechrest et al, 1972). First, the ing items and multiplying the sum by 2. The overall score is
SCS (Neff, 2003b) was translated from English to Korean the sum of the 3 subscales. Lee et al. (2019) translated the
by a translator, and subsequently back-translated to Eng- measure to Korean and examined its psychometric properties
lish by another translator. Then, a group of three other in a sample of 430 adults from community health care cent-
translators compared the original and back-translated ver- ers (nonclinical subgroup) and 50 adults from community
sions to evaluate concept equivalence. Any discrepancies mental health centers (clinical subgroup). The results dem-
between the two versions were addressed by retranslation onstrated good internal consistency indicated by Cronbach’s
of that item by the group of translators until consensus alpha (α) values of 0.81 for Depression, 0.84 for Anxiety,
was achieved. This third step was repeated with a new and 0.85 for Stress subscales. In the current study, McDon-
group of translators until both versions were judged to ald’s omega (ω) was 0.90 for Depression, 0.86 for Anxi-
have the same content. The third step of this sequence is ety, and 0.89 for Stress subscales; Cronbach’s alpha (α) was
referred to as the decentering procedure, which focuses 0.90 for Depression, 0.85 for Anxiety, and 0.89 for Stress
on the meaning and context of the instruments rather than subscales.
on verbatim literal translation (Brislin, 1970; Werner &
Campbell, 1970).
Happiness

Measures The Subjective Happiness Scale (SHS; Lyubomirsky & Lep-

per, 1999) is a 4-item survey that assesses global subjective
Participants completed all measures in Korean including the happiness. Items are rated on a 7-point scale (1= not a very
K-SCS, which was translated for the current study. All the happy person, 7 = a very happy person). The total score is
other Korean versions of the measures were translated, vali- obtained by calculating the average of the items after reverse
dated, and widely used in Korean populations. coding the fourth item (e.g., 7 = 1, 6 = 2), with higher scores
indicating greater self-reported happiness. A Korean version
of the measure has been widely used with Korean popula-
Self‑Compassion tions including adolescents (Kim, 2009), college students
(Jang, 2016), and older adults (Jo & Song, 2012). In the
The resulting retranslation was entitled the Korean Self-Com- current study, McDonald’s ω was 0.83 and Cronbach’s α
passion Scale (K-SCS). Following the original SCS (Neff, was 0.81.
2003b), the K-SCS contains 26 items consisting of 6 subscales:
self-kindness, mindfulness, common humanity, self-judgment,
over-identification, and isolation. The items are rated on a Life Satisfaction
5-point scale (1 = almost never, 5 = almost always), and items
pertaining to self-judgment, over-identification, and isolation The Satisfaction with Life Scale (SWLS; Diener et al., 1985)
subscales are reverse coded. The newly translated version of is a 5-item survey that assesses subjective judgement of
the K-SCS constructed for this study was used (see Supple- one’s life satisfaction. Items are rated on a 7-point scale (1 =
mentary Information). Items are averaged for each subscale, strongly disagree, 7 = strongly agree). Items are summed to
and subscale scores are summed to yield an overall score, with obtain an overall score, with higher scores reflecting greater
higher scores reflecting greater self-compassion. In a study life satisfaction. A Korean version of the Satisfaction with
with 232 undergraduate students in the United States (Neff, Life Scale (Cho & Cha, 1998) has been used with adoles-
2003b), psychometric properties for the SCS demonstrates cents, young adults, and police officers (Lim, 2012) as well
good internal consistency (Cronbach’s alpha = 0.94) and test- as with geriatric populations (Seo et al., 2013). Results from
retest reliability (r = 0.94). In the current study, both McDon- the current study demonstrated McDonald’s ω of 0.91 and
ald’s omega (ω) and Cronbach’s alpha (α) were 0.92. Cronbach’s α of 0.90.
 Mindfulness

Data Analyses 2014; Karakasidou et al., 2017). In order to examine whether

there was a significant difference between male and female
The analyses were conducted with Mplus 8 (Muthén & groups on the SCS scores, an independent t-test was con-
Muthén, 2017) for model analyses and SPSS 29.0 for ducted. Results indicated a significant difference between the
descriptive statistics and validity analyses. For all the mod- two groups (t (508) = 2.85, p = 0.005), with males reporting
els, weighted least squares mean and variance-adjusted esti- slightly higher self-compassion (M = 3.30, SD = 0.53) than
mator (WLSMV), as suggested by previous studies (e.g., females (M = 3.17, SD = 0.52). However, the effect size
Finney & DiStefano, 2006) for ordered-categorical items (Cohen’s d = 0.25) indicated a small to medium practical
with five or fewer response options. The current study significance of this difference.
assessed three corresponding CFA and ESEM models to Model fit indices for each model are illustrated in Table 1.
investigate the construct-relevant dimensionality of the Both the 2-factor CFA and ESEM models demonstrated poor
K-SCS. These models include the following: (1) 2-factor model fit, indicating that the K-SCS cannot be represented
model with 2 unitary factors representing self-compassion by self-compassion and self-coldness subscales. The CFA
and self-coldness, (2) 6-factor correlated model with six model for the 6-factor correlated structure illustrated ade-
components of self-compassion, and (3) single-bifactor quate CFI and TLI values but did not show an acceptable
model with a general self-compassion factor and 6 specific RMSEA value. On the other hand, the 6-factor correlated
factors. ESEM approach indicated excellent model fit, with good
In the CFA models, items were permitted to load only to moderate standardized factor loadings (λ = 0.34 to 0.99,
onto their a priori target factors with cross-loadings being M = 0.65; see Table 2). Similarly, while the CFA for the
constrained to zero. In the ESEM models, items were single-bifactor model showed poor fit, the ESEM approach
allowed to load onto non-target factors. For the ESEM indicated excellent fit. The examination of the parameter
model pertaining to (1) and (2), the model was estimated estimates (Table 3) indicated that the general self-com-
with target rotation, while the ESEM model pertaining to passion factor was well-defined with mostly strong (i.e., λ
(3) was specified as orthogonal to the general factor. For > 0.54) and significant factor loadings (λ = 0.10 to 0.76,
the evaluation of model fit, typical goodness-of-fit indices M = 0.57). Most of the specific factors were well defined,
(Hu & Bentler, 1999; Marsh et al., 2005) were used, which with the majority of factor loadings greater than 0.32: self-
include the comparative fit index (CFI), the Tucker-Lewis judgment (λ = 0.46 to 0.69, M = 0.52), common humanity
Index (TLI), and the root mean square error of approxima- (λ = 0.28 to 0.41, M = 0.52), isolation (λ = 0.39 to 0.67,
tion (RMSEA) with its 90% confidence interval (90% CI). M = 0.54), mindfulness (λ = 0.36 to 0.54, M = 0.46), and
For CFI and TLI, values above 0.90 were considered ade- over-identification (λ = 0.46 to 0.56, M = 0.50). However,
quate and 0.95 as excellent; for RMSEA, values below 0.08 self-kindness was weakly defined (λ = 0.05 to 0.59, M =
were considered adequate and those below 0.06 as excellent. 0.30), indicating that the items pertaining to self-kindness
For the evaluation of the magnitude of the factor loadings, might be indicators of self-compassion rather than the spe-
values above 0.71 were considered excellent, values between cific component of self-compassion. These findings are con-
0.63 and 0.70 as very good, values between 0.55 and 0.62 as sistent with previous studies, such as those by Neff et al.
good, values between 0.33 and 0.54 as fair, and values below
0.32 as poor (Comrey & Lee, 2013).
For the analyses of convergent and discriminant validity, Table 1  Goodness-of-fit indices for the models
the New K-SCS was examined in relation to other related Models Model Fit Indices
constructs such as negative emotions (i.e., stress, depression,
CFI TLI RMSEA 90% CI
and anxiety), happiness, and life satisfaction. For the evalu-
ation of the magnitude of the correlations, values between 2-factor correlated CFA 0.88 0.87 0.11 0.10 – 0.11
± 0.50 and ± 1 were considered strong, values between ± 2-factor correlated ESEM 0.87 0.84 0.11 0.11 – 0.12
0.30 and ± 0.49 as medium, and values between ± 0.29 and 6-factor correlated CFA 0.91 0.90 0.09 0.09 – 0.10
± 0.01 as small correlations. 6-factor correlated ESEM 0.98 0.96 0.06 0.06 – 0.06
Bifactor-CFA 0.84 0.80 0.13 0.12 – 0.13
Bifactor-ESEM 0.99 0.97 0.05 0.04 – 0.06
Results Note. CFA confirmatory factor analysis, ESEM exploratory structural
equation modeling, CFI comparative fit index, TLI Tucker-Lewis
The descriptive statistics from the New K-SCS yielded a index; RMSEA root mean square error of approximation
total SCS mean score of 19.39 (SD = 3.19), which is similar
to the mean scores and standard deviation values reported in (2019) and Tóth-Király et al. (2017), which also observed
the previous validation studies (e.g., Garcia-Campayo et al., reduced factor loadings with the ESEM approach. These
Mindfulness

Table 2  Standardized factor Items CFA ESEM

loadings for the 6-factor models
a
SF(λ) SK(λ) SJ(λ) CH(λ) IS(λ) MI(λ) OI(λ)

Self-kindness
sk5 0.77 0.84 -0.15 0.03 -0.02 -0.08 0.12
sk12 0.83 0.92 -0.07 0.00 0.02 -0.14 0.13
sk29 0.69 0.54 0.14 0.14 0.00 -0.05 0.05
sk23 0.74 0.52 0.19 -0.05 0.05 0.33 -0.14
sk26 0.73 0.42 0.28 0.07 -0.02 0.34 -0.16
Self-judgment
sj1 0.75 0.15 0.54 0.02 0.00 0.06 0.21
sj8 0.28 0.07 0.66 0.03 -0.09 -0.25 -0.02
sj11 0.82 0.04 0.59 -0.03 0.22 0.21 0.04
sj16 0.83 -0.10 0.53 0.02 0.09 0.18 0.32
sj21 -0.08 -0.01 0.57 0.02 -0.06 -0.42 -0.18
Common Humanity
ch3 0.83 0.17 0.06 0.38 0.12 0.24 -0.03
ch7 0.75 0.00 0.05 0.85 -0.06 -0.03 -0.02
ch10 0.77 -0.13 -0.05 0.99 -0.05 -0.01 0.03
ch15 0.76 0.17 -0.00 0.51 0.14 0.10 -0.13
Isolation
is4 0.79 -0.05 0.15 0.09 0.45 0.02 0.25
is13 0.81 -0.03 -0.03 -0.01 0.95 -0.07 -0.03
is18 0.80 0.04 -0.03 -0.03 0.99 -0.06 -0.13
is25 0.87 -0.01 0.04 0.05 0.58 0.05 0.25
Mindfulness
mi9 0.79 0.18 0.01 0.09 -0.04 0.64 0.05
mi14 0.83 0.04 -0.06 0.15 0.07 0.72 0.08
mi17 0.80 0.07 -0.08 0.12 0.03 0.67 0.13
mi22 0.83 0.32 -0.12 0.25 0.08 0.34 0.04
Over-identification
oi2 0.85 0.15 0.10 -0.03 0.14 0.01 0.62
oi6 0.83 -0.03 0.22 0.12 0.21 -0.11 0.56
oi20 0.80 0.04 0.06 -0.09 0.03 0.13 0.76
oi24 0.80 0.07 0.05 0.00 0.06 0.11 0.67

Note. CFA confirmatory factor analysis, ESEM exploratory structural equation modeling, SF specific factor,
SK self-kindness, SJ self-judgment, CH common humanity, IS isolation, M mindfulness, OI over-identifica-
tion, λ standardized factor loadings
Target factor loadings are in bold. Each item loaded on its respective specific factor, while cross-loadings
were constrained to zero

previous studies explained that with the ESEM approach, score, items from the self-judgment, isolation, and over-
the factor loading of each item is less significant when the identification subscales are reverse-coded (e.g., 5 = 1, 4 =
general factor (i.e., self-compassion) is well-defined. Over- 2) to reflect a self-compassionate attitude. This means that,
all, the results support the use of 6-factor correlated ESEM once these items are reverse-coded, lower scores on these
and bifactor ESEM approaches when using the new K-SCS. subscales indicate greater self-compassion. However, when
This indicates that both the 6 subscale scores as well as the exploring the relationship between each subscale and other
overall total score can be used when using the K-SCS. related psychological variables, the self-judgment, isolation,
To evaluate the validity of the new K-SCS, Pearson cor- and over-identification subscale scores were not reverse-coded,
relation analyses were conducted to explore the relationship in order to reflect the actual levels of self-judgment, isola-
between self-compassion and other related psychological tion, and over-identification tendencies. As a result, reverse
variables (Table 4). Typically, when obtaining a total SCS coding was only applied to these subscales when calculating
 Mindfulness

Table 3  Standardized factor Items Bifactor CFA Bifactor ESEM

loadings for bifactor models
GF(λ) SF(λ) GF(λ) SK(λ) SJ(λ) CH(λ) IS(λ) MI(λ) OI(λ)

Self-kindness
sk5 0.57 0.56 0.58 0.59 -0.07 0.14 -0.07 0.10 -0.05
sk12 0.63 0.64 0.67 0.55 -0.01 0.12 -0.06 -0.01 -0.04
sk29 0.55 0.38 0.58 0.31 0.10 0.16 -0.06 -0.04 -0.05
sk23 0.61 0.33 0.72 0.09 -0.07 -0.01 -0.17 -0.03 -0.19
sk26 0.60 0.29 0.76 -0.05 -0.04 0.03 -0.28 -0.12 -0.26
Self-judgment
sj1 0.60 0.44 0.56 0.03 0.46 -0.05 0.06 -0.01 0.17
sj8 0.12 0.70 0.10 0.10 0.69 0.00 0.06 -0.13 0.02
sj11 0.66 0.43 0.62 -0.11 0.46 -0.11 0.16 0.00 0.09
sj16 0.69 0.40 0.56 -0.12 0.47 -0.11 0.17 0.08 0.30
sj21 -0.21 0.63 -0.15 -0.00 0.53 0.01 0.02 -0.38 -0.08
Common Humanity
ch3 0.63 0.33 0.64 0.06 -0.05 0.28 -0.01 0.12 -0.09
ch7 0.45 0.68 0.46 0.13 -0.01 0.65 -0.06 0.05 -0.11
ch10 0.46 0.77 0.47 0.07 -0.10 0.75 -0.05 0.07 -0.07
ch15 0.53 0.47 0.54 0.12 -0.07 0.41 0.00 0.08 -0.17
Isolation
is4 0.66 0.33 0.55 -0.04 0.21 -0.01 0.39 0.02 0.27
is13 0.59 0.66 0.52 -0.05 0.09 -0.05 0.67 -0.07 0.14
is18 0.60 0.62 0.55 -0.03 0.08 -0.04 0.67 -0.09 0.06
is25 0.73 0.38 0.64 -0.07 0.08 -0.04 0.44 -0.03 0.27
Mindfulness
mi9 0.60 0.52 0.65 0.00 -0.12 0.04 -0.13 0.42 -0.06
mi14 0.62 0.64 0.67 -0.05 -0.14 0.06 -0.04 0.54 -0.03
mi17 0.60 0.58 0.65 -0.02 -0.16 0.05 -0.06 0.50 0.01
mi22 0.66 0.36 0.63 0.23 -0.11 0.22 -0.01 0.36 -0.08
Over-identification
oi2 0.71 0.45 0.66 -0.00 0.10 -0.12 0.18 -0.09 0.47
oi6 0.71 0.37 0.58 -0.04 0.26 -0.02 0.27 -0.12 0.46
oi20 0.63 0.59 0.60 -0.07 0.06 -0.19 0.13 0.02 0.56
oi24 0.66 0.46 0.61 -0.02 0.06 -0.10 0.14 0.04 0.49

Note. CFA confirmatory factor analysis, ESEM exploratory structural equation modeling, SF specific factor,
SK self-kindness, SJ self-judgment, CH common humanity, IS isolation, MI mindfulness, OI over-identifi-
cation, λ standardized factor loadings
Target factor loadings are in bold. Each item loaded on its respective specific factor, while cross-loadings
were constrained to zero.

the overall self-compassionate attitude score, in order to align overall SCS were negatively associated with stress, depres-
with the direction of self-compassion. Results indicated that sion, and anxiety, while self-judgment, isolation, and over-
the new K-SCS has satisfactory construct validity. Specifically, identification subscales were negatively related to happiness
the new K-SCS demonstrated good convergent validity, where and life satisfaction.
self-kindness, mindfulness, and common humanity subscales
as well as the overall SCS were positively correlated with
happiness and life satisfaction; self-judgment, isolation, and Discussion
over-identification subscales were positively correlated with
stress, depression, and anxiety subscales. The new K-SCS The current study developed and validated a newly translated
also demonstrated good discriminant validity, where self- Korean version of the SCS (the K-SCS) and examined its
kindness, mindfulness, and common humanity as well as the psychometric properties and factor structure. Specifically,
Mindfulness

Table 4  Correlations of self-compassion subscales, self-compassion supported the 2-bifactor model, which conceptualizes the
total score, and mental health indices SCS as a scale that broadly measures compassionate attitude
Measures STR DEP ANX SWLS SHS towards oneself via self-kindness, mindfulness, and common
humanity subscales and uncompassionate attitude towards
SK -0.38** -0.44** -0.28** 0.39** 0.46**
oneself via self-judgment, over-identification, and isolation
CH -0.23** -0.34** -0.19** 0.37** 0.41**
subscales. The lack of support for the single bifactor model
MI -0.36** -0.39** -0.25** 0.35** 0.38*
may be due to the difference in the targeted population (i.e.,
SJ 0.42** 0.41** 0.33** -0.32** -0.42**
middle school students in Park (2023) study and adults in
IS 0.51** 0.55** 0.44** -0.50** -0.56**
the current study) and/or differences in the translation of the
OI 0.54** 0.50** 0.36** -0.42** -0.47**
items. Further efforts should be made to conduct validation
SCS -0.57** -0.61** -0.43** 0.55** 0.62**
studies of the SCS across populations.
Note. SK self-kindness, CH common humanity, MI mindfulness, SJ Using the newly translated K-SCS upon feedback ses-
self-judgment, IS isolation, OI over-identification, SCS total self-com- sions, findings from the current study indicate that the
passion, STR stress, DEP depression, ANX anxiety, SWLS satisfaction K-SCS can be used to assess the 6 subscales as well as the
with life, SHS happiness
overall total score of the SCS. The present study’s support
p < 0.05 *; p < 0.001**
for the ESEM approach further contributes to the stead-
ily increasing literature on ESEM and on research of the
the current study examined the factor structure and the SCS using ESEM (e.g., Neff et al., 2019; Tóth-Király
validity of the K-SCS using CFA, ESEM, and correlation et al., 2017). However, while the 6-factor correlated ESEM
analyses. Findings from the current study supported the use demonstrated adequate to excellent factor loadings for all
of 6-factor correlated and bifactor models using an ESEM items, some of the items in the bifactor ESEM resulted in
approach. Findings further support previous studies that less well-defined factor loadings. Particularly, Items 23 and
indicated that the Korean version of the SCS is most appro- 26 resulted in weak factor loadings, which are both under
priately analyzed using a 6-factor model (Jin & Lee, 2009; the self-kindness factor. According to Morin et al. (2016),
Jo & Kim, 2011; Ku & Jung, 2016; Kim & Kim, 2019; Lee when using an ESEM approach, not all specific factors are
& Lee, 2010). However, some previous studies (Park et al., required to be equally well-defined if the global factor is
2020; Shin, 2018) reported different findings. For instance, well-defined. The current study further supports this argu-
Shin (2018) tested eight models (i.e., the single factor model, ment, where the general factor (i.e., self-compassion) and
2-factor model, 4-factor model, 6-factor model, single bifac- most of the specific factors were well-defined. This finding is
tor model, and other expanded versions of the 4-factor mod- also consistent with previous studies (e.g., Neff et al., 2019;
els) and supported the use of a dual-factor model which Tóth-Király et al., 2017) where some of the specific items
supports the 6-factor structure as well as the self-coldness were reported to be less-defined when accounting for the
general factor for over-identification, self-judgment, and iso- general self-compassion factor (i.e., global factor).
lation domain-specific factors. On the other hand, Park et al. The findings on self-kindness could possibly be explained
(2020) supported the 6-factor structure as well as two gen- in the context of Korean culture. Items in the self-kindness
eral factors (i.e., self-compassion for self-kindness, mindful- subscale includes two main ways: (1) showing love and a
ness, and common humanity domain specific factors; self- caring attitude towards oneself when faced with difficulties
coldness for over-identification, self-judgment, and isolation or when in suffering and (2) showing understanding and
domain specific factors). These findings indicate that there patience despite one’s flaws, shortcomings, and failures. The
is inconsistency regarding comprising the Korean SCS as a latter type of self-kindness, arguably, may be perceived dif-
tool that measures a construct (i.e., self-compassion) or two ferently for South Koreans, whose cultural values are deeply
constructs (i.e., self-compassion and self-coldness). Given rooted in Confucian values of good conduct, hard work, and
that both studies used the translated version by Kim et al. self-cultivation and improvement. The items that resulted in
(2008), which was observed to include less accurately trans- acceptable to good factor loadings in the self-kindness sub-
lated items, it is less clear whether these supported models scale are related to showing care, love, and kindness to one-
indicate the most accurate representation of the factor struc- self, while the other 2 items with weak factor loadings high-
ture for the Korean version of the SCS. Another study that light being tolerant and patient with one’s flaws, failures, and
attempted to translate and validate the measure with Korean challenges. In Korean culture, examining and identifying
middle school students also indicated partially different find- one’s own personal shortcomings are important Confucian
ings (Park, 2023) from the current study. Park (2023) con- values concerning work ethic and personal growth (DeVos,
ducted CFA and ESEM to explore 6 different factor models 1998). It is possible that for Koreans, being tolerant of one’s
and supported the 6-factor structure, which is consistent with flaws and being patient despite one’s challenging life circum-
findings from the current study. However, Park (2023) also stances may be more associated with Confucian values of
 Mindfulness

drive for self-correction and self-improvement rather than explore the factor structure of the K-SCS. Findings sup-
with self-kindness. ported the use of the bifactor ESEM approach, an advanced
statistical method that has been recommended to address
Limitations, Future Directions, and Conclusions the highly restrictive nature of the CFA when exploring the
SCS (Neff, 2016a). This indicates that the K-SCS can be
The current study has added to the growing body of litera- used by scoring and interpreting items by 6 subscales as
ture on the importance of reviewing and assessing translated well as by summing up the mean of each subscale score for
measures, examination of the factor structure of translated a total score. By providing an improved Korean version of
measures, and cultural considerations when translating and the SCS, the present study may facilitate further research
analyzing the psychometric properties of translated meas- on the SCS for Korean population and highlight the impor-
ures. With a growing effort to study the role of self-com- tance of culturally appropriate translations when exploring
passion across various cultural groups, the SCS has been translated measures.
translated and validated in multiple languages. However, However, the current study should be examined within
most studies either briefly described the translation process the context of the study’s limitations. First, although there
or used the previously translated version without assessing are advantages to using online survey platforms such as
the accuracy of the translated measure. The current study expedited data collection, cost-effectiveness, and increased
highlights the consideration of the translation equivalence diversity (Buhrmester et al., 2011; Goodman et al., 2013;
prior to conducting a validation study for the SCS. The cur- Peer et al., 2017), a number of researchers have raised con-
rent study was the first to revisit the translational equivalence cerns regarding the quality of data obtained from online
of the Korean version of the SCS (Kim et al., 2008) and survey portals (Chmielewski & Kucker, 2020; Smith et al.,
retranslate the scale to accurately capture the intended mean- 2016). Follmer et al. (2017) indicated that the demograph-
ing and connotation of the original SCS prior to examining ics of individuals who participate in online survey studies
the factor structure of the measure. are skewed towards certain groups of people, which may
In the context of South Korea, it was observed that most lead to conclusions that are less representative of the general
studies have used the SCS translated by Kim et al. (2008) population. Other studies have identified inattentiveness as
to continue exploring the factor structure of the SCS and a major concern that may threaten the validity of results
assess the validity of the measure for Korean population. (Aruguete et al., 2019; Buhrmester et al., 2018). However,
Although it is important to assess the most appropriate way a recent meta-analysis of 90 online survey studies indicated
of using and interpreting the SCS for the Korean population, that data obtained from online platforms yielded psychomet-
it is imperative to create an accurately translated measure ric properties and validity indicators comparable to those
prior to examining the psychometric properties of the meas- collected from conventional sources such as college students
ure. The current study observed misleading translations in (Walter et al., 2019). These studies indicate that the findings
the SCS translated by Kim et al. (2008), which may have from the current study warrant further research on K-SCS
interfered with measuring the intended construct. This raises with more diverse populations including college population,
concerns about the validity of the Korean version of the clinical population, and groups with a wide range of socio-
SCS, even before examining its factor structure. The current economic status. Furthermore, given the inconsistent find-
study shows that reviewing the translation of the materials is ings in the factor structure of the K-SCS, it is important for
an essential preliminary step to exploring the psychometric the future studies to conduct validation studies of the K-SCS
properties of the scale. This also has implications for the using the identical measure. Second, in addition to exploring
broader field of self-compassion research, as several trans- the factor structure of the scale, measurement invariance
lated versions have been introduced since the introduction should also be conducted across various groups (e.g., across
of the SCS (Neff, 2003b). As the field continues to explore gender, age, culture) in order to draw a more solid conclu-
the SCS across various cultural groups, it is indispensable to sion about the use of the SCS in various comparative studies
investigate and interpret both the universality of self-com- across diverse cultural groups. Lastly, other analyses such
passion and the extent of cultural differences in its manifes- as Rasch analysis should be considered to robustly consider
tations. In order to do so, studies should first ensure that the differences in individual’s responses that may stem from dif-
translation of the measure is accurate and culturally appro- ferences in the emphasis of the content in a specific item as a
priate to the target population. This comprehensive approach result of translations (Medvedev & Krägeloh, 2022).
will enhance our understanding of self-compassion, ensuring
Supplementary Information The online version contains supplemen-
that measures are both culturally sensitive and universally tary material available at https://​doi.​org/​10.​1007/​s12671-​024-​02453-z.
applicable.
Overall, findings from the current study provided a new Acknowledgements We are also thankful to our participants and to
version of the K-SCS and examined different models to translators who made this research possible.
Mindfulness

Author Contributions Si Woo Chae: Conceptualization, Formal analy- Psychological and Personality Science, 11(4), 464–473. https://​
sis, Methodology, Writing – Original draft doi.​org/​10.​1177/​19485​50619​875149
Jeong Eun Cheon: Data curation, Investigation, writing – review Cho, M. H., & Cha, K. H. (1998). Comparative analyses of well-
& editing being across countries. Jipmoondang.
Janet Latner: Writing – review & editing, Supervision Comrey, A. L., & Lee, H. B. (2013). A first course in factor analysis.
Yong-Hoon Kim: Funding Acquisition, Project administration, Psychology press.
Resources Costa, J., Marôco, J., Pinto-Gouveia, J., Ferreira, C., & Castilho, P.
(2015). Validation of the psychometric properties of the Self-
Funding This research was supported by the Yonsei Signature Compassion Scale. Testing the factorial and factorial invariance
Research Cluster Program of 2024 (2024-22-0166). of the measure among borderline personality disorder, anxiety
disorder, eating disorder and general populations. Clinical Psy-
Data Availability The data that support the findings of this study are chology and Psychotherapy, 23(5), 460–468. https://​doi.​org/​10.​
available from the corresponding author, Si Woo Chae, upon reason- 1002/​cpp.​1974
able request. DeVos, G. A. (1998). Confucian family socialization: The religion,
morality, and aesthetics of propriety. Confucianism and the
Declarations Family, 329–380.
Diedrich, A., Hofmann, S. G., Cuijpers, P., & Berking, M. (2016).
Ethics Approval The study procedures including the informed consent Self-compassion enhances the efficacy of explicit cognitive
were approved by Yonsei University Institutional Review Board office reappraisal as an emotion regulation strategy in individuals with
(IRB number: 7001988-202201-HR-1079-07). major depressive disorder. Behaviour Research and Therapy,
82, 1–10. https://​doi.​org/​10.​1016/j.​brat.​2016.​04.​003
Informed Consent Informed Consent was obtained via online survey Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The
portal from all participants before the start of the survey. Satisfaction With Life Scale. Journal of Personality Assessment,
49(1), 71–75. https://​doi.​org/​10.​1207/​s1532​7752j​pa4901_​13
Use of the Artificial Intelligence The current study did not use any form Finney, S. J., & DiStefano, C. (2013). Nonnormal and categorical
of artificial intelligence tools. data in structural equation modeling. In G. R. Hancock & R. O.
Mueller (Eds.), Structural equation modeling: A second course
(2nd ed., pp. 439–492). IAP Information Age Publishing.
Conflict of Interests The authors declared no potential conflicts of in- Follmer, D. J., Sperling, R. A., & Suen, H. K. (2017). The role of
terest with respect to the research, authorship, and/or publication of MTurk in education research: Advantages, issues, and future
this article. directions. Educational Researcher, 46(6), 329–334. https://​doi.​
org/​10.​3102/​00131​89X17​725519
Garcia-Campayo, J., Navarro-Gil, M., Andrés, E., Montero-Marin,
J., López-Artal, L., & Demarzo, M. M. P. (2014). Validation
References of the Spanish versions of the long (26 items) and short (12
items) forms of the Self-Compassion Scale (SCS). Health and
Anggraeni, D. T., & Kurniawan, I. N. (2012). Self-compassion and Quality of Life Outcomes, 12(1), 4. https://​d oi.​o rg/​1 0.​1 186/​
satisfaction with life: A preliminary study on Indonesian college 1477-​7525-​12-4
students. Jurnal Psikologi Universitas Islam Indonesia, 53(23), Gilbert, P. (2005). Social mentalities: A biopsychosocial and evolu-
105–108. tionary approach to social relationships. In M. W. Baldwin (Ed.),
Arimitsu, K., & Hofmann, S. G. (2015). Cognitions as mediators in Interpersonal cognition (pp. 299–333). The Guilford Press.
the relationship between self-compassion and affect. Personality Gilbert, P., & Procter, S. (2006). Compassionate mind training for
and Individual Differences, 74, 41–48. https://​doi.​org/​10.​1016/j.​ people with high shame and self-criticism: Overview and pilot
paid.​2014.​10.​008 study of a group therapy approach. Clinical Psychology & Psycho-
Aruguete, M. S., Huynh, H., Browne, B. L., Jurs, B., Flint, E., & therapy, 13(6), 353–379. https://​doi.​org/​10.​1002/​cpp.​507
McCutcheon, L. E. (2019). How serious is the “carelessness” Gilbert, P., McEwan, K., Matos, M., & Rivis, A. (2011). Fears of com-
problem on Mechanical Turk? International Journal of Social passion: Development of three self-report measures. Psychology
Research Methodology, 22(5), 441–449. https://​doi.​org/​10.​1080/​ and Psychotherapy, 84(3), 239–255. https://​doi.​org/​10.​1348/​
13645​579.​2018.​15639​66 14760​8310X​526511
Brislin, R. W. (1970). Back-translation for cross-cultural research. Goodman, J. K., Cryder, C. E., & Cheema, A. (2013). Data collection
Journal of Cross-Cultural Psychology, 1(3), 185–216. https://​ in a flat world: The strengths and weaknesses of Mechanical Turk
doi.​org/​10.​1177/​13591​04570​00100​301 samples. Journal of Behavioral Decision Making, 26(3), 213–224.
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s https://​doi.​org/​10.​1002/​bdm.​1753
Mechanical Turk: A new source of inexpensive, yet high-quality, Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covari-
data? Perspectives on Psychological Science, 6(1), 3–5. https://​ ance structure analysis: Conventional criteria versus new alterna-
doi.​org/​10.​1177/​17456​91610​393980 tives. Structural Equation Modeling, 6(1), 1–55. https://​doi.​org/​
Buhrmester, M., Talaifar, S., & Gosling, S. (2018). An evaluation of 10.​1080/​10705​51990​95401​18
Amazon’s Mechanical Turk, its rapid rise, and its effective use. Hupfeld, J., & Ruffieux, N. (2011). Validierung einer deutschen ver-
Perspectives on Psychological Science, 13(2), 149–154. https://​ sion der Self-Compassion Scale (SCS-D). Zeitschrift für Klinische
doi.​org/​10.​1177/​17456​91617​706516 Psychologie und Psychotherapie, 40(2), 115–123. https://​doi.​org/​
Chen, J., Yan, L. S., & Zhou, L. H. (2011). Reliability and validity of 10.​1026/​1616-​3443/​a0000​88
Chinese version of Self-Compassion Scale. Chinese Journal of Jang, H. J. (2016). Comparative study of health promoting lifestyle
Clinical Psychology, 19(6), 734–736. profiles and subjective happiness in nursing and non-nursing
Chmielewski, M., & Kucker, S. C. (2020). An MTurk crisis? students. Advances in Science and Technology Letters, 128(7),
Shifts in data quality and the impact on study results. Social 78–82. https://​doi.​org/​10.​14257/​astl.​2016.​128.​16
 Mindfulness

Jin, H. J., & Lee, K. H. (2009). The validation of the Korean version of Springer International Publishing. https://​doi.​org/​10.​1007/​978-3-​
Self-Compassion Scale. The Korea Journal of Youth Counseling, 030-​77644-2_​4-1
17(2), 1–20. https://​doi.​org/​10.​35151/​kyci.​2009.​17.2.​001 Morin, A. J. S., Arens, A. K., & Marsh, H. W. (2016). A bifactor
Jo, S. J., & Kim, J. W. (2011). Preliminary study for the development exploratory structural equation modeling framework for the iden-
of the Korean version of Self-Compassion Scale. Korean Journal tification of distinct sources of construct-relevant psychometric
of Psychosomatic Medicine, 19(1), 34–40. multidimensionality. Structural Equation Modeling, 23(1), 116–
Jo, K. H., & Song, B. S. (2012). Effect of family cohesion, subjective 139. https://​doi.​org/​10.​1080/​10705​511.​2014.​961800
happiness and other factors on death anxiety in Korean elders. Muris, P., & Otgaar, H. (2020). The process of science: A critical
Journal of Korean Academy of Nursing, 42(5), 680–688. https://​ evaluation of more than 15 years of research on self-compassion
doi.​org/​10.​4040/​jkan.​2012.​42.5.​680 with the Self-Compassion Scale. Mindfulness, 11(6), 1469–1482.
Karakasidou, E., Pezirkianidis, C., Galanakis, M., & Stalikas, A. https://​doi.​org/​10.​1007/​s12671-​020-​01363-0
(2017). Validity, reliability and factorial structure of the Self Muris, P., van den Broek, M., Otgaar, H., Oudenhoven, I., & Lennartz,
Compassion Scale in the Greek population. Journal of Psychol- J. (2018). Good and bad sides of self-compassion: A face valid-
ogy and Psychotherapy, 7(313), 2161–0487. https://​doi.​org/​10.​ ity check of the Self-Compassion Scale and an investigation of
4172/​2161-​0487.​10003​13 its relations to coping and emotional symptoms in non-clinical
Kim, C. S. (2009). A study of determinants of happiness on adoles- adolescents. Journal of Child and Family Studies, 27, 2411–2421.
cents. Korean Journal of Health Psychology, 14(3), 649–665. https://​doi.​org/​10.​1007/​s10826-​018-​1099-z
Kim, H. J., & Kim, M. (2019). Validation study of the Korean ver- Muthén, L. K., & Muthén, B. O. (2017). Mplus user’s guide: Statisti-
sion of self-compassion scale-focusing on correctional work- cal analysis with latent variables (8th ed.). Muthén & Muthén.
ers. Journal of the Korea Academia-Industrial Cooperation Neff, K. D. (2003a). Self-compassion: An alternative conceptualiza-
Society, 20(8), 321–330. https://​doi.​org/​10.​5762/​KAIS.​2019.​ tion of a healthy attitude toward oneself. Self and Identity, 2(2),
20.8.​321 85–101. https://​doi.​org/​10.​1080/​15298​86030​9032
Kim, K. E., Yi, G. D., Cho, Y. R., Chai, S. H., & Lee, W. K. (2008). Neff, K. D. (2003b). The development and validation of a scale to
The validation study of the Korean version of the Self-Compas- measure self-compassion. Self and Identity, 2, 223–250. https://​
sion Scale. Korean Journal of Health Psychology, 13(4), 1023– doi.​org/​10.​1080/​15298​86030​9027
1044. https://​doi.​org/​10.​17315/​kjhp.​2008.​13.4.​012 Neff, K. D. (2009). The role of self-compassion in development: A
Ku, D. Y., & Jung, M. C. (2016). Study of factor validity of Korean healthier way to relate to oneself. Human Development, 52(4),
version Self-Compassion. The Journal of the Korea Contents 211. https://​doi.​org/​10.​1159/​00021​5071
Association, 16(9), 160–169. Neff, K. D. (2016a). The self-compassion scale is a valid and theo-
Lee, E. H., Moon, S. H., Cho, M. S., Park, E. S., Kim, S. Y., Han, J. S., retically coherent measure of self-compassion. Mindfulness, 7(1),
& Cheio, J. H. (2019). The 21-item and 12-item versions of the 264–274. https://​doi.​org/​10.​1007/​s12671-​015-​0479-3
Depression Anxiety Stress Scales: Psychometric valuation in a Neff, K. D. (2016b). Does self-compassion entail reduced self-judg-
Korean population. Asian Nursing Research, 13(1), 30–37. https://​ ment, isolation, and over-identification? A response to Muris,
doi.​org/​10.​1016/j.​anr.​2018.​11.​006 Otgaar, and Petrocchi (2016). Mindfulness, 7(3), 791–797. https://​
Lee, W. K., & Lee, K. (2010). The validation study of the Korean doi.​org/​10.​1007/​s12671-​016-​0531-y
version of Self-Compassion Scale with adult women in commu- Neff, K. D., Rude, S. S., & Kirkpatrick, K. L. (2007). An examination
nity. Journal of Korean Neuropsychiatric Association, 193–200. of self-compassion in relation to positive psychological function-
Lim, Y. J. (2012). Psychometric properties of the satisfaction with life ing and personality traits. Journal of Research in Personality,
scale among Korean police officers, university students, and ado- 41(4), 908–916. https://​doi.​org/​10.​1016/j.​jrp.​2006.​08.​002
lescents. Korean Journal of Psychology: General, 31(3), 877–896. Neff, K. D., Whittaker, T. A., & Karl, A. (2017). Examining the factor
López, A., Sanderman, R., Smink, A., Zhang, Y., van Sonderen, E., structure of the Self-Compassion Scale in four distinct popula-
Ranchor, A., & Schroevers, M. J. (2015). A Reconsideration of tions: Is the use of a total scale score justified? Journal of Person-
the Self-Compassion Scale’s total score: Self-compassion versus ality Assessment, 99(6), 596–607. https://​doi.​org/​10.​1080/​00223​
self-criticism. PloS One, 10(7), e0132940. https://​doi.​org/​10.​ 891.​2016.​12693​34
1371/​journ​al.​pone.​01329​40 Neff, K. D., Tóth-Király, I., Yarnell, L. M., Arimitsu, K., Castilho, P.,
Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative Ghorbani, N., Guo, H. X., Hirsch, J. K., Hupfeld, J., Hutz, C. S.,
emotional states: Comparison of the Depression Anxiety Stress Kotsou, I., Lee, W. K., Montero-Marin, J., Sirois, F. M., de Souza,
Scales (DASS) with the Beck Depression and Anxiety Inventories. L. K., Svendsen, J. L., Wilkinson, R. B., & Mantzios, M. (2019).
Behaviour Research and Therapy, 33(3), 335–343. https://d​ oi.o​ rg/​ Examining the factor structure of the Self-Compassion Scale in 20
10.​1016/​0005-​7967(94)​00075-U diverse samples: Support for use of a total score and six subscale
Lyubomirsky, S., & Lepper, H. S. (1999). A measure of subjective scores. Psychological Assessment, 31(1), 27–45. https://​doi.​org/​
happiness: Preliminary reliability and construct validation. Social 10.​1037/​pas00​00629
Indicators Research, 46(2), 137–155. https://​doi.​org/​10.​1023/A:​ Park, N. W. (2023). The structural validity of the Self-Compassion
10068​24100​041 Scale for the Korean youth. Korean Journal of Youth Studies,
Macbeth, A., & Gumley, A. (2012). Exploring compassion: A meta- 30(7), 119–146. https://d​ oi.o​ rg/1​ 0.2​ 1509/K
​ JYS.2​ 023.0​ 7.3​ 0.7.1​ 19
analysis of the association between self-compassion and psycho- Park, H., Kim, E., & Shin, H. (2020). Examining factor structure of the
pathology. Clinical Psychology Review, 32(6), 545–552. https://​ Korean version of the Self-Compassion Scale: Distinctive effect of
doi.​org/​10.​1016/j.​cpr.​2012.​06.​003 self-compassion and self-coldness. Korea Journal of Counseling,
Marsh, H. W., Hau, K. T., & Grayson, D. (2005). Goodness of fit in 21(3), 227–251.
structural equation modeling. In A. Maydeu-Olivares & J. J. Peer, E., Brandimarte, L., Samat, S., & Acquisti, A. (2017). Beyond
McArdle (Eds.), Contemporary psychometrics (pp. 275–340). the Turk: Alternative platforms for crowdsourcing behavioral
Lawrence Erlbaum Associates. research. Journal of Experimental Social Psychology, 70, 153–
Medvedev, O. N., & Krägeloh, C. U. (2022). Rasch measurement 163. https://​doi.​org/​10.​1016/j.​jesp.​2017.​01.​006
model. In O. N. Medvedev, C. U. Krägeloh, R. J. Siegert, & N. N. Petrocchi, N., Ottaviani, C., & Couyoumdjian, A. (2014). Dimensional-
Singh (Eds.), Handbook of assessment in mindfulness research. ity of self-compassion: Translation and construct validation of the
Mindfulness

self-compassion scale in an Italian sample. Journal of Mental Health, Tóth-Király, I., Bőthe, B., & Orosz, G. (2017). Exploratory struc-
23(2), 72–77. https://​doi.​org/​10.​3109/​09638​237.​2013.​841869 tural equation modeling analysis of the Self-Compassion
Reise, S. P., Moore, T. M., & Haviland, M. G. (2010). Bifactor models Scale. Mindfulness, 8(4), 881–892. https://​d oi.​o rg/​1 0.​1 007/​
and rotations: Exploring the extent to which multidimensional data s12671-​016-​0662-1
yield univocal scale scores. Journal of Personality Assessment, Walter, S. L., Seibert, S. E., Goering, D., & O’Boyle, E. H. (2019). A
92(6), 544–559. https://​doi.​org/​10.​1080/​00223​891.​2010.​496477 tale of two sample sources: Do results from online panel data and
Sechrest, L., Fay, T. L., & Zaidi, S. H. (1972). Problems of translation conventional data converge? Journal of Business and Psychol-
in cross-cultural research. Journal of Cross-Cultural Psychology, ogy, 34(4), 425–452. https://​doi.​org/​10.​1007/​s10869-​018-​9552-y
3(1), 41–56. https://​doi.​org/​10.​1177/​00220​22172​00300​103 Werner, O., & Campbell, D. T. (1970). Translating, working through
Seo, S., Cho, M., Kim, Y., & Ahn, J. (2013). The relationships among interpreters, and the problem of decentering. In R. Naroll & R.
satisfaction with food-related life, depression, isolation, social Cohen (Eds.), A handbook of method in cultural anthropology (pp.
support, and overall satisfaction of life in elderly South Koreans. 398–420). American Museum of Natural History.
Journal of the Korean Dietetic Association, 19(2), 159–172. Williams, M. J., Dalgleish, T., Karl, A., & Kuyken, W. (2014). Examin-
Shin, N. Y. (2018). The factor structure of the Korean Self-Compassion ing the factor structures of the five facet mindfulness questionnaire
Scale (K-SCS): A study of university students. Journal of Human and the Self-Compassion Scale. Psychological Assessment, 26(2),
Understanding and Counseling, 39, 73–85. 407–418. https://​doi.​org/​10.​1037/​a0035​566
Sinaiko, H. W., & Brislin, R. W. (1973). Evaluating language trans-
lations: Experiments on three assessment methods. Journal of Publisher's Note Springer Nature remains neutral with regard to
Applied Psychology, 57(3), 328. https://d​ oi.o​ rg/1​ 0.1​ 037/h​ 00346​ 77 jurisdictional claims in published maps and institutional affiliations.
Smith, G. T., McCarthy, D. M., & Zapolski, T. C. (2009). On the
value of homogeneous constructs for construct validation, theory Springer Nature or its licensor (e.g. a society or other partner) holds
testing, and the description of psychopathology. Psychological exclusive rights to this article under a publishing agreement with the
Assessment, 21(3), 272. https://​doi.​org/​10.​1037/​a0016​699 author(s) or other rightsholder(s); author self-archiving of the accepted
Smith, G. T., McCarthy, D. M., & Zapolski, T. C. (2009). On the manuscript version of this article is solely governed by the terms of
value of homogeneous constructs for construct validation, theory such publishing agreement and applicable law.
testing, and the description of psychopathology. Psychological
Assessment, 21(3), 272–284. https://​doi.​org/​10.​1037/​a0016​699

View publication stats