Previous research has confirmed that self-compassion plays a positive role in helping people cope with the frustrations and pains of daily life (Ewert et al., 2021) and is also important in organizational life (Janssen et al., 2018). Self-compassion is a fundamental concept in positive psychology, which is grounded in Buddhist psychology’s notion of “compassion.” The construct of self-compassion (SC) was introduced, conceptualized, and operationalized by Kristin D. Neff. Self-compassion comprises three bi-polar and mutually interacting dimensions: self-kindness versus self-judgment; common humanity versus isolation; and mindfulness versus over-identification (Neff, 2003). Compassion for oneself is considered a healthy, positive self-concept. In recent years, researchers have found that self-compassion can effectively enhance an individual’s intrinsic motivation, reduce negative emotions and improve the ability to adjust positively to adverse events (Stutts et al., 2018; Andersson et al., 2021; Hughes et al., 2021). Self-compassion is associated with various positive psychological outcomes, including greater mental resilience (Kotera et al., 2021a,b), higher levels of well-being (Passmore et al., 2018), and the ability to enhance individual levels of self-compassion through appropriate interventions (Wilson et al., 2019).

Self-compassion also plays a vital role in organizational life, where suffering is an unavoidable part (Dutton et al., 2014). At work, employees frequently experience unpleasant feelings and sensations due to various workplace occurrences, such as workplace, including workplace restrictions (Yu et al., 2021), toxic interactions with coworkers (Quade et al., 2019), and failed entrepreneurial endeavors (Patzelt et al., 2021). It is also possible for pain to arise from unrelated events and spread into the workplace (Eby et al., 2016), such as employment-related difficulties (Oka et al., 2017), past traumatic experiences (Izutsu et al., 2008), and global pandemics (Guler et al., 2021). Notably, emerging organizational studies have started to investigate how practicing self-compassion in the face of adversity can positively affect organizational outcomes. Researchers have documented that individuals high in self-compassion can effectively handle difficult situations and cope better with burnout (Terry et al., 2013). Meanwhile, researchers have also found that self-compassion could buffer the negative relationship between work–family conflict and psychological well-being (Rafique et al., 2018; Yang et al., 2022).

Countries have recently conducted research on self-compassion in the workplace, exploring its role in facilitating and protecting employees’ responses to setbacks and challenges. For example, a study based on a European professional group found that mindfulness training increased employees’ self-compassion, which in turn reduced their stress and fatigue (van der Meulen et al., 2021); a cross-sectional study in Japan found that employee mental health problems were significantly and negatively associated with levels of self-compassion (Kotera et al., 2022); a study based on a U.S. occupational group found that employees with a self-compassionate mindset were better able to cope with feelings of job isolation and depression (Andel et al., 2021). Although evidence of the positive role of self-compassion in the workplace has been obtained from studies in both eastern and western cultural contexts, however, research in Chinese occupational groups is underrepresented. This is reflected in the inadequacy of both the research instrument and the research findings. As the Eastern country with the most significant number of labor resources, exploring the manifestation and influence mechanism of self-compassion in Chinese occupational groups is undoubtedly indispensable to enrich the empirical research results of self-compassion and deepen the theoretical understanding of self-compassion in the Eastern cultural context.

Valid and reliable instruments are used to measure self-compassion in research. The Self-Compassion Scale (SCS; Neff, 2003) and its short-form variant (SCS-SF; Raes et al., 2011) are the most commonly used scales measuring six different aspects of self-compassion, which includes self-kindness, self-judgment, common humanity, isolation, mindfulness, and over-identification. Nevertheless, some researchers argue that empirical studies do not always support the six-factor model. Instead, they suggest that the SCS and SCS-SF only represent two aspects: self-criticism and self-compassion (Muris et al., 2016; Babenko and Guo, 2019; Halamová et al., 2021). To address this, Strauss et al. (2016) integrated different definitions and measures of self-compassion from existing research, suggesting that self-compassion is a cognitive, emotional, and behavioral dynamic process that involves (1) recognizing suffering, (2) understanding that suffering is a part of every human experience, (3) feeling empathy and relating to people who are suffering, (4) tolerating unpleasant feelings caused by the sufferer, and (5) taking action to ease discomfort. Based on the five-factor structure described above, Gu et al. (2020) developed the Sussex-Oxford Compassion for the Self (SOCS-S) and the Sussex-Oxford Compassion for Others Scale (SOCS-O). Both scales consist of five elements that constitute overall self-compassion and show good reliability and validity with their original samples.

Based on the above theoretical conceptualization and scale development, the SOCS-S has been shown to be a reliable and valid tool for measuring self-compassion in various countries and industries (Kim and Seo, 2021; de Krijger et al., 2022; Lucarini et al., 2022; Shreffler et al., 2022). However, to further validate its use, it is necessary to examine its measurement invariance across gender and culture and to measure self-compassion levels in different occupational groups in various countries. To the best of our knowledge, there is no validation study of SOCS-S in Chinese to date.

Therefore, the purpose of this study is twofold. First, we aim to develop a Chinese version of the SOCS-S and evaluate its psychometric properties by administering it to a Chinese occupational group. We used classical test theory (CTT), which includes internal consistency, factor analysis, and measurement invariance across gender. Internal consistency assessed the reliability of the SOCS-S scale without retesting. Factor analysis examined whether the same factor structure would be found in a Chinese sample. Measurement invariance cross-validated the five-factor model across gender groups. We also utilized modern approaches such as item response theory (IRT), also known as graded response model analysis (GRM), and SOCS-S network analysis. Self-compassion at work refers to the extent to which employed individuals are well prepared (cognitively and emotionally) to effectively handle the difficulties faced in their organizations while on the job. We are interested in understanding how much self-compassion Chinese employees require to respond using each scale category, and how effectively each item differentiates (the α parameter) between employees with varying degrees of self-compassion (the b parameters). Furthermore, we conducted a network analysis to examine the strength, proximity, and betweenness indices of every node included in the estimated network. The traditional approach to psychometric measurement of scales is to conduct factor analysis, mainly focusing on the differences in the included items (Sellbom and Tellegen, 2019). However, this traditional method only calculates the correlation of the included items. Network analysis presents the characteristics and information of a scale in the form of a network consisting of nodes and edges, to visualize the structure and information of a scale by means of the recurrence of these relations, and to describe and explain the scale from the perspective of the network (Contreras et al., 2019). Simultaneously, it confirms the feasibility of SOCS-S as a measurement tool in Chinese occupational groups. By measuring the level of self-compassion in Chinese occupational groups and exploring the adaptability of the SOCS-S in different cultural contexts, we provide instrumental support for further research related to self-compassion in the workplace in China.

The purpose of this study was to assess the psychometric properties of the Sussex Oxford Compassion for the Self Scale (SOCS-S) in a Chinese occupational group. The SOCS-S was translated and administered to the Chinese occupational group to overcame the limitations of previous self-compassion scales, specifically its good content validity in ensuring that the items are relevant to suffering and not overlapping with other constructs. Thus, the scale helps us in understanding the level of self-compassion in the Chinese occupational group. Our findings support the factor structure of the SOCS-S and its robust psychometric properties.

Consistent with a previous study (Gu et al., 2020), this research found high internal consistency for the SOCS-S, as evidenced by McDonald’s ω (0.819–0.936) and Cronbach’s α (0.706–0.927). The five-factor structure showed reasonable model fit for the scale. Multi-group CFA analyses validated the measurement invariance across gender for the SOCS-S, indicating that males and females had similar interpretations of the item content. These findings are consistent with earlier observations (Gu et al., 2020). However, our results differ from Kim and Seo (2021) study, which failed to find scalar measure invariance. Other studies, such as de Krijger et al. (2022), only found metric invariance. This indicates the need for further research on the measurement invariance of the SOCS-S across gender, as the correlations between the five characteristics of self-directed compassion in Chinese working populations may differ from those in other cultures. One possible explanation for these cultural differences in self-compassion is that it is a social mindset that can be influenced by various social environments (Gilbert, 2014). In China, gender equality is an important part of education (Zhu et al., 2018). Since the founding of the People’s Republic of China, the reform of institutions, laws and policies has promoted the popularization of gender equality concepts and the de-gendering of social roles. Therefore, when facing pain and setbacks in work and life, both men and women have similar coping strategies. Furthermore, “home culture” is a unique cultural concept of Chinese enterprises, which originates from the profound understanding and importance of “home” among Chinese people (Huang et al., 2022). In the Chinese perspective, “home” not only refers to a family or a clan, but also encompasses the nation and the world. Therefore, enterprises should be united and harmonious like a big family, and develop together. “Persistence” and “self” are two important elements of home culture (Dang and Zhao, 2020). Only by pursuing goals persistently can the team improve its execution and efficiency; only by establishing correct self-awareness and values can the team maintain its personality and creativity.”

This study confirmed that the SOCS-S met a strict assumption of unidimensionality (Morizot et al., 2007) with: (1) a ratio greater than 3 between the first and second eigenvalues; and (2) a dominant factor explaining at least 20% of the variance. The GRM analysis revealed that the SOCS-S is a psychometrically sound instrument, as the slopes and thresholds provided informative indicators of individuals’ self-compassion levels. The ICC curves show that item 5 (i.e., “I try to make myself feel better when I’m distressed, even if I cannot do anything about the cause”), item 10 (i.e., “When I’m going through a difficult time, I try to look after myself”), item 15 (i.e., “When I’m upset, I try to do what’s best for myself”), and item 20 (i.e., “When I’m upset, I do my best to take care of myself”) exhibit higher discrimination and are more useful in assessing the latent trait related to taking action or being motivated to alleviate suffering. In contrast, item 2 (i.e., “I understand that everyone experiences suffering at some point in their lives”), item 7 (i.e., “I understand that feeling upset at times is part of human nature”), item 12 (i.e., “Like me, I know that other people also experience struggles in life”), and item 17 (i.e., “I know that we can all feel distressed when things do not go well in our lives”) exhibit lower discrimination and are less useful in evaluating the latent trait. These four items are associated with understanding the universality of suffering. According to these findings, the items related to AM were able to differentiate employees based on their levels of self-compassion, while the items related to US were not. According to Strauss et al. (2016) review, the definition of compassion often includes related terms such as empathy, which are used to define each other. It is illuminating to consider the overlap and differences between these terms. One of the key distinctions between compassion and empathy is that acting or the desire to act to alleviate suffering is viewed as a core feature of compassion, but not empathy. The Total Information Curve (TIC) supports the overall effectiveness of the SOCS-S, demonstrating precise estimation of scores across an extensive range of the continuum with a low standard error of estimation (SEE).

Theoretical definitions of compassion describe it as a cognitive, affective, and behavioral process consisting of five elements: (a) recognizing suffering, (b) understanding that suffering is a part of every human experience, (c) feeling empathy and relating to those who suffer, (d) tolerating unpleasant feelings caused by suffering, and (e) taking action to ease discomfort (Strauss et al., 2016). Nonetheless, the theory of self-compassion attempts to operationalize a multidimensional description by consolidating various conceptualizations. Our findings suggest an insufficient balance of item difficulty across distinct domains of self-compassion operationalized by the SOCS-S. Seven items (Item 1, 2, 6, 7, 12, 16, 17) had values of b₁ (corresponding to the lowest response category) below −3. Additionally, the parameters b1 to b3 were predominantly negative, indicating that individuals low on self-compassion tended to respond to SOCS-S items with high response categories (3–5) suggesting that the items are formulated too “easy.” Moreover, all SOCS-S items employ negative phrasing, such as sadness, fear, anger, frustration, guilt, shame, etc., which may also contribute to social desirability biases. Negative phrasing alone may not be sufficient to create difficulty in selecting items that reflect self-compassion, as items that express other socially desirable personality traits, such as self-esteem, self-pity, self-forgiveness, empathy, etc., may also be well-received.

Network analysis showed that item 20 (i.e., “When I’m upset, I do my best to take care of myself”) and item 10 (i.e., “When I’m going through a difficult time, I try to look after myself”) had the relatively higher indicators of centrality: strength, closeness, and betweenness, while item 1 (i.e., “I’m good at recognizing when I’m feeling distressed”) had the lowest centrality indicator values. Therefore, the results acquired in the network analysis are comparable to those discovered in the GRM study. Moreover, the items that scored the highest on indices of centrality in network analysis also showed themselves to be highly informative for the latent construct, as was brought to light by GRM.

The contributions of this study are threefold. First, this is the very first validation study of the SOCS-S conducted among employed individuals in China. Previous validation studies using modified versions of the SOCS-S were conducted in Western countries (de Krijger et al., 2022; Lucarini et al., 2022) and Asian cultures (Kim and Seo, 2021). Second, classical tests and modern methods were used to calculate the reliability. More specifically, the McDonald’s ω coefficient and Cronbach’s α were computed in CTT. The graphs of IIF and TIF marginal reliability were calculated using a modern methodology. While the reliability of the CTT only shows the mean value of all items, IRT has the benefit of offering a bottom-up view of reliability through the IIF graph. More precisely, it records different changes according to the pattern of each item. Hence, this study offers a more practical approach to the psychometric qualities of the SOCS-S. Third, this is the first study to apply network analysis experimental method to the SOCS-S. By employing IRT and network analysis, we were able to identify the most difficult, discriminant, and central items on the SOCS-S.

There are certain limitations to the generalizability of these findings. First, the use of self-reported measures introduces potential biases such as social desirability and recall biases, highlighting the need for further longitudinal studies or the use of non-self-report measures to triangulate the SOCS-S results, which would be useful for future studies. Second, while self-compassion has been studied as an underlying trait that affects individuals’ perceptions and outcomes in organizations, recent attention in management suggests that practicing self-compassion can lead to happier, more successful, and resilient employees (Chen, 2018). Therefore, future research should investigate the relationships between the aforementioned constructs, including the antecedents and outcomes of self-compassion, using the SOCS-S. Third, research in psychology, pedagogy, and counseling suggests that self-compassion can uniquely contribute to positive responses to challenges and is a source of comfort (Neff, 2003; Neff et al., 2007; Neff and Vonk, 2009; Ma et al., 2022). Therefore, future studies could explore the positive impact of self-compassion across various industries to deepen our understanding. Fourth, the SOCS-S requires further validation as some psychometric properties, such as test–retest reliability, and convergent and discriminant validity, were not assessed in this study due to its scope.

In general, the evidence from this study utilizing advanced psychometric testing (e.g., IRT and network analysis) confirmed that the SOCS-S possesses strong homogeneity, discriminative power, and excellent reliability. As a result, the SOCS-S is a robust measurement tool that can be effectively employed in both practical and research settings to assess levels of self-compassion among individuals in the Chinese workplace.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Local Ethics Committee at Tianjin Normal University. The patients/participants provided their written informed consent to participate in this study.

HM and XW conceptualized the manuscript, drafted the manuscript, and conducted all the statistical analyses. HL revised the manuscript. HM collected the data. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyg.2023.1110076/full#supplementary-material