The Effect of Self-Compassion on Academic Burnout in Undergraduate Students

Background and Aim of Study: Due to high academic demands, undergraduate students are prone to academic burnout. Academic burnout is emotional exhaustion caused by the pressures of academic tasks. Having self-compassion can be a protective factor against academic burnout. When dealing with numerous academic pressures and demands, undergraduate students with high self-compassion will be able to see themselves and their situations positively, resulting in lower academic burnout. The aim of the study: to investigate the effect of self-compassion on academic burnout in undergraduate students in Indonesia. Material and Methods: This study was conducted using the quantitative approach with a non-experimental design. The Maslach Burnout Inventory-Student Survey (MBI-SS) was used to measure academic burnout (α=0.824), and the measurement of self - compassion used the Self-Compassion Scale (SCS) (α=0.878), which has been adapted into the Indonesian language. A sample of undergraduate students in Indonesia between the ages of 17 and 25 was taken using the convenience sampling technique. This study used simple linear regression analysis. Results: The regression analysis from data of 474 undergraduate students shows that there is a negative and significant effect of self-compassion on academic burnout in undergraduate students (R 2 =0.076, F=38.960, p<0.001). Furthermore, the variable of gender and a current semester each has a significant effect on academic burnout, but the finding reveals the opposite with the age variable. Conclusions: Self-compassion has a negative effect on academic burnout in undergraduate students. This explains that an increase in self-compassion will be followed by reduction in academic burnout in undergraduate students.


Introduction
During college, undergraduate students will face a variety of challenges, such as academic, social, and personal challenges (Aguayo et al., 2019). These challenges include fitting into a new social environment, being away from home, and adjusting to the different class schedule. In academic settings, undergraduate students must dedicate additional time outside of class hours to research study material and complete the coursework, both individually and in groups. As a result of the numerous academic obligations they face, undergraduate students are prone to burnout, which is defined by Yavuz and Dogan (2014) as a feeling of exhaustion mentally and physically. The longitudinal study done by Ohio State University in August 2020 and April 2021 discovered an increase in the number of undergraduate students feeling burnout. The percentage increase was 31%, whereas in August 2020, 40% of undergraduate students experienced burnout, but 71% did in April 2021 (Citroner, 2021). Furthermore, LM Psychology UGM (2021) research for the representation of academic burnout on 300 undergraduate students of the Faculty of Socio-Humanities UGM found that 73.12% of them experienced high fatigue (exhaustion), 57.03% experienced high cynicism and indifference to tasks (cynicism), and 49.43% experienced a decrease in selfconfidence in academic tasks (reduced self-efficacy). According to Schaufeli et al. (2002, as cited in Oyoo et al., 2018, exposure to physically and emotionally demanding tasks over an extensive period can cause burnout. Initially, people in helping professions such as teachers, psychologists, doctors, and nurses were more likely to experience burnout. Undergraduate students have the potential to experience burnout because their primary academic pursuits can be viewed as "work" (J. Lee et al., 2010). Thus, the term burnout can be applied to students, called academic burnout. A variety of external and internal factors can contribute to academic burnout. According to Lin and Yang (2021), the main external factor that causes academic burnout is high-demand and high-pressure tasks. The lack of support from parents, lecturers, and peer; poor parenting styles, and low socioeconomic status are a few additional factors that might contribute to academic burnout. According to Charkhabi et al. (2014), one of the factors contributing to academic burnout is a lack of resources available to accomplish academic work. Lack of resources could lead to stress, which can lower motivation and hinder undergraduate students' ability to finish activities. Therefore, Charkhabi et al. (2014) concluded that when universities place high academic demands but provide low resources, the likelihood of academic burnout increases. The personality, self-esteem, and attribution style of an individual are internal factors that contribute to academic burnout. These three factors have an impact on how one responds to negative emotions and circumstances: neurotic personality trait causes the tendency to use maladaptive strategies, including engaging in self-blame behavior; low self-esteem leads to burnout, which is characterized by exhaustion and demotivation; and stagnant attributional style leads to a resistance to change in the future and in turn, a passive attitude appears, causing the individuals to become more cynical, unmotivated, and emotionally charged (Lin & Yang, 2021). Self-compassion can reduce the potential of burnout because it entails the ability to handle and appraise negative emotions and situations more positively and more effectively (K. Lee & Lee, 2020;Neff, 2003). Therefore, it is stated that self-compassion can prevent academic burnout due to high academic demands (K. Lee & Lee, 2020). Self-compassion has been described as an attitude of openness, caring, and understanding on a cognitive and emotional level toward one's shortcomings, sufferings, and failures, as one would treat others during hard times (Elices et al., 2017;K. Lee & Lee, 2020;Neff, 2003). Self-compassion can be applied in various contexts, including in the educational setting. The qualitative study (Lockard et al., 2014) shows that undergraduate students with high self-compassion experience lesser negative emotions, deal with their problems in a more rational and accepting manner, and at the same time assume responsibility for their problems. According to Zhang et al. (2016), undergraduate students can also respond to academic failure by seeking help and support from others. They take failure as an opportunity to improve and avoid making the same mistakes (Neff, 2003;Smeets et al., 2014). The authors of the study (Pypenko et al., 2020) found that direct teacher-student contact has a positive impact on the physical and mental health of higher education stakeholders. The construct of self-compassion consists of three components, each with polar opposites: self-kindness vs. self-judgment; common humanity vs. isolation; and mindfulness vs. over-identification (Neff, 2003). Individuals with adequate self-compassion can understand that suffering and failure are unpleasant (mindfulness) without reducing the kindness toward oneself (self-kindness) because it is a part of the human condition (common humanity), thus enabling them to appropriately cope with those suffering and failure. Conversely, individuals who judge themselves (selfjudgment), isolate themselves from others (isolation) and overreact (over-identification) in the face of suffering and/or failure are said to have inadequate selfcompassion (Neff, 2003). Dreisoerner et al. (2021) found that self-judgment leads to procrastination, rumination, and a generally lower level of selfimprovement. Isolation, on the other hand, leads to stress and burnout (Dreisoerner et al., 2021;Neff, 2003). With self-compassion, individuals perceive negative emotions more positively and take a more rational approach, which allows them to cope effectively (Neff, 2003). If left untreated, academic burnout may negatively impact one's academic achievement; academic interest; and interactions between undergraduate students and the academic community, including lecturers, staff, and other students (Charkhabi et al., 2014;Rahmatpour et International Journal of Science Annals, Vol. 6, No. 1, 2023 рrint ISSN: 2617-2682; online ISSN: 2707-3637; DOI:10.26697/ijsa al., 2019). According to one study on undergraduate students in Iran by Rahmati (2015), prolonged academic burnout causes stress, anxiety, frustration, depression, and low problem-solving skills. In academic contexts, several studies have found that undergraduate students with high levels of selfcompassion experience lower academic burnout (Beaumont et al., 2016a;2016b;K. Lee & Lee, 2020). Self-compassion in undergraduate students is also a protective factor against chronic academic stress (Zhang et al., 2016), since this causes undergraduate students to perceive academic demands as opportunities for selfdevelopment, and to remain treating themselves well, even under academic pressures (Beaumont et al., 2016a;2016b;K. Lee & Lee, 2020;Zhang et al., 2016). So far, there has been limited study into self-compassion and its relationship to academic burnout. Most of the literature view self-compassion as a moderating variable, such as a protective factor of emotional responses to chronic academic stress (Zhang et al., 2016); a moderator between academic demands, academic burnout, and depression (K. Lee & Lee, 2020); and a moderator between perceived stress and academic burnout (Paramadina, 2022). Additionally, Sabharwal et al. (2021) looked into the moderating role of academic self-efficacy in the relationship between self-compassion and academic burnout. Some studies on self-compassion and academic burnout are also constrained by sample sizes. Beaumont et al. (2016b) studied self-compassion and burnout in final-year Cognitive Behavioral Psychotherapist (CBP) undergraduate students and Person-Centered Counsellors undergraduate students with a small sample size (n=54). In another study, Beaumont et al. (2016a) has its limitation on its small sample (n=103) which were midwifery undergraduate students, exclusively from one university in the North West of England, United Kingdom. Among the limited number of research discussing selfcompassion and academic burnout (Paramadina, 2022), none directly seeks to determine the effect of selfcompassion on academic burnout. The aim of the study. To investigate the effect of selfcompassion on academic burnout in undergraduate students in Indonesia to improve on the findings and to remove some limitations faced by prior studies.

Materials and Methods
In this study, non-experimental quantitative approach is used since the researchers wish to understand the causal relationship without administering any direct manipulation (Gravetter & Forzano, 2018). In this study, two variables -academic burnout and self-compassion are used. The population of this study is Indonesian undergraduate students. According to the Higher Education Statistics from the Directorate General of Higher Education of Indonesia (2020), there were 7,113,663 undergraduate students enrolled in 2020. The number of samples to be taken is at least 349 undergraduate students when utilizing the reference table of Isaac and Michael at the error level of 5% (Sugiyono, 2013). Undergraduate students between the ages of 17 and 25 enrolled in universities in Indonesia constitute the research sample. The sampling method used is convenience sampling (Gravetter & Forzano, 2018), which includes any person who meets the requirements, is the simplest to locate, and is willing to participate in the research as a respondent.

Measuring Instruments
The Skala Welas Diri (SWD) (Sugianto et al., 2020), adapted from the Self-Compassion Scale (SCS) by Neff (2003), was used to measure self-compassion. The instrument has three factors with respective polar opposites, which are self-kindness vs. self-judgment, common humanity vs. isolation, and mindfulness vs. over-identification. The SWD operationalizes the selfcompassion dimensions into six components comprised of 26 Likert-scale items, ranging from 1 (almost never) to 5 (almost always). Psychometric assessment of SWD by Sugianto et al. (2020) using the construct validity tests, Confirmatory Factor Analysis (CFA), and Cronbach's Alpha coefficients yielded a valid and reliable instrument, and showed that SWD has the same theoretical model as SCS. Before proceeding into the measuring process using SWD, SWD underwent a readability test on four individuals to ensure the items are comprehensible. Researchers also ran through a psychometric assessment of SWD, which confirmed the reliability (α=0.878) and construct validity (corrected item total correlation >0.20) of the instrument. To generate the scale score and component score, the average score is determined by dividing the total score by the number of items, producing a score ranging from 1.00 to 5.00, which signifies the level of selfcompassion one has. Researchers used the Maslach Burnout Inventory-Student Survey (MBI-SS), which has been adapted and translated into Indonesian by Maharani (2019). The reliability testing shows that MBI-SS are reliable with α=0.932 (Maharani, 2019). MBI-SS has 15 items consisting of three dimensions, i.e. exhaustion, cynicism, and efficacy, using the Likert scale with a range of 1 (very unlikely) to 5 (very likely). The greater the individual's academic burnout, the higher the total score. Before implementing the MBI-SS, researchers conducted readability tests on four people, with the conclusion that all of the items could be understood. Researchers' reliability testing results demonstrate that the MBI-SS is reliable (α=0.824) and valid (corrected item total correlation >0.20). The overall score, which demonstrates the degree of academic burnout experienced, is calculated to obtain a scale score and dimension score.

Research Procedure
Since the data will be collected virtually and in person, researchers will create the questionnaires in both computerized and printed versions. In addition to the SWD and MBI-SS instruments, researchers will inquire demographic background in the questionnaire such as gender, age, originating university, current major, and current semester. For virtual data collection, researchers International Journal of Science Annals, Vol. 6, No. 1, 2023 рrint ISSN: 2617-2682; online ISSN: 2707-3637; DOI:10.26697/ijsa will use Google Forms and share the computer-based questionnaire links on various online social media platforms, such as WhatsApp, Instagram, Telegram, Line, etc. For the person-to-person data collection, the paper-and-pencil questionnaires will be printed out and distributed directly by researchers to undergraduate students several Indonesian universities. The data collection process is planned to be carried out in September 2022 -February 2023. This study was tested using simple linear regression analysis in the JASP 0.17.0.0 software. Prior to the regression test, researchers ran through a set of assumption tests (Field, 2018), consisting of normality distributed errors, linearity, independent error, and homoscedasticity.

Results
The period of data collection is September 2022 to February 2023. In this study, 474 college students between the ages of 17 and 25 (M=20.25, SD=1.52) participated as respondents. Table 1 provides an overview of the characteristics of the respondents. The majority of respondents are female (85.65%), aged between 20-22 years (60.13%), are undergoing semester 5 (16.24%), studying at universities located in Jabodetabek (54.85%), and studying humanities (33.76%). Table 2 displays the descriptive statistics of the selfcompassion and academic burnout variables, alongside the dimensions of each variable. The total score of selfcompassion ranges from 1.3-4.5 (M=3.22; SD=0.52) and the self-compassion component score ranges from 1.0-5.0. The total score of academic burnout ranges from 16.0-17.30 (M=45.71; SD=9.06) and its dimensions, denote the score variance from the lowest to the highest. T-test is conducted based on gender for self-compassion and academic burnout. Since the data did not pass the normality assumption test despite passing the homogeneity of variance assumption test, researchers used Mann-Whitney U Test (Table 2). There is no significant difference in self-compassion between female undergraduate students (M=3.22, SD=0.53) and male undergraduate students (M=3.19, SD=0.47), U=14.722, p>0.05. There is no significant difference in the entirety of the self-compassion components between male and female undergraduate students. Conversely, academic burnout is shown to be significantly higher in female undergraduate students (M=46.11, SD=8.98) compared to male undergraduate students (M=43.34, SD=9.27), U=16566.5, p<0.05. The same result is shown in the Exhaustion and Cynicism dimensions, in which the score of the female undergraduate students is significantly higher than male undergraduate students, with U=16875.0 and U=16011.0, respectively. Only the Efficacy dimension shows no significant difference between female undergraduate students (M=18.68, SD=5.72) and male undergraduate students (M=18.85, SD=6.34), U=13470.0, p>0.05.  Note. a Mann-Whitney test between the score of male and female, b n=406, c n=68. *p<0.05, **p<0.01, ***p<0.001.
The regression results for predicting academic burnout are shown in Table 3. Academic burnout is significantly and negatively influenced by self-compassion (R 2 =0. 076, F=38.960, p<0.001). Academic burnout will decrease by 4.770 points for every additional 1 score of self-compassion. 7.6% of self-compassion can predict academic burnout, while 92.4% was predicted by other variables. In accordance with Gravetter and Forzano (2018), self-compassion has a small effect on academic burnout, as shown by its R 2 of 0.076.
By adding the age and semester variables as predictors along with self-compassion, all three significantly affect academic burnout, R 2 =0.084, F=15.409, p<0.001. The addition of these two predictors increases R 2 by 0.008, although the effect is still small. Self-compassion, semester, and age together can predict academic burnout by 8.4%. Self-compassion and age have negative effect on academic burnout (b=-4.899 and -0.848, respectively), while the semester has a positive effect (b=0.654). This shows that increasing self-compassion and age causes a decrease in academic burnout while increasing semesters will cause an increase in academic burnout. Although these three predictors had a significant effect, self-compassion has the most effect (β=-0.284).
A fourth predictor, gender, was added to the regression analysis. These four predictors together have a significant effect on academic burnout, R 2 =0.100, F=13.073, p<0.001. An R 2 of 0.100 indicates that these four predictors together have a moderate effect (Gravetter & Forzano, 2018). As a result, female undergraduate students will likely experience greater academic burnout than male undergraduate students. The age variable, however, emerged as the only predictor from this regression analysis that did not have a significant effect. Table 4 displays the results of regression test with the self-compassion components as a predictor.
The six components simultaneously has a significant effect on academic burnout, R 2 =0.146, F=14.518, p<0.001. When compared to the overall effect of selfcompassion (R 2 =0.076), the effect of self-compassion components are higher and can be categorized as moderate (Gravetter & Forzano, 2018). Only three components, which are Self-Kindness, Mindfulness, and Over-identification, that respectively has a significant effect on academic burnout, where the Mindfulness component has the highest effect (β=-0.252). If the predictors are added with the variables of age, semester, and gender, R 2 increases to 0.176 (F=10.984, p<0.001). As stated in the previous results, only the Self-kindness, Mindfulness, and Over-identification components, and the semester and gender variables have significant effects.

Discussion
This study aims to seek the effect of self-compassion on academic burnout in undergraduate students. Results show that there is a negative significant effect of selfcompassion on academic burnout, indicating that the higher the self-compassion one has, the lower the academic burnout one experiences. The same conclusion is found across research, such as in a study on final-year Cognitive Behavioral Psychotherapist (CBP) undergraduate students and Person-Centered Counsellors undergraduate students (Beaumont et al., 2016b), on midwifery undergraduate students (Beaumont et al., 2016a), and on undergraduate students of counseling classes in Seoul Cyber University (Kyeong, 2013). Individuals with high self-compassion will be more resilient in the face of negative situations, allowing them to become more compassionate and more accepting of their shortcomings and failures, avoid engaging in selfcriticism, and overcome existing challenges in a positive and constructive way. High self-compassion also enables individuals to be able to see and appraise their situations as an opportunity for growth (K. Lee & Lee, 2020;Neff, 2003). When encountering a lot of pressure and demands as undergraduate students, individuals with high selfcompassion will be able to see themselves and their situations positively, which leads to lower academic burnout. They will also be able to use emotion-focused coping strategies that reduce fear of failure and negative thoughts (Cheraghian et al., 2016). Academic burnout is positively influenced by the number of semesters students have completed. This finding indicates that academic burnout increases as a student's time in college lengthens. This condition is understandable given that undergraduate students face more academic fatigue as they take on more challenging coursework and are required to finish their final project or thesis during their final semester. This finding is supported by Salmela-Aro and Upadyaya (Doğru & Kabasakal, 2023), which stated that exhaustion happens during the first year of college study, and escalates into desensitization during the next years. Additionally, Ramirez-Asis et al. (2023) noted that as the number of semesters increase, undergraduate students become more emotionally exhausted and cynical about their academic careers. This also explains why there are only two aspects that significantly differ between male and female undergraduate students: exhaustion and cynicism. This study also found that age does not have an effect on academic burnout, reiterating the research findings of Aguayo et al. (2019) in students of University of Granada, and García-Izquierdo et al. (2018) in midwifery students of University of Murcia. Researchers suggest that this is due to the shared characteristics of respondents, in which the majority of them are young adults between 17 and 25 years old (Santrock, 2011). Female undergraduate students experience greater academic burnout than male undergraduate students International Journal of Science Annals, Vol. 6, No. 1, 2023 рrint ISSN: 2617-2682; online ISSN: 2707-3637; DOI:10.26697/ijsa when compared by gender. This result can be explained because female students show higher stress levels than male undergraduate students (Castellanos, 2018). Compared to male undergraduate students, female undergraduate students are more likely to employ coping mechanisms that emphasize emotions (Aguayo et al., 2019). Female undergraduate students employ the emotion-focused coping technique to get through the feelings that come with burnout. This condition also explains why the exhaustion and cynicism dimensions in female undergraduate students are higher than in male undergraduate students. This is supported by the findings of a study conducted on medical undergraduate students at the University of Liege, Belgium who found female undergraduate students had higher scores in emotional exhaustion (Kilic et al., 2021). Self-kindness, mindfulness, and over-identification each have a significant effect on academic burnout, in line with a study on undergraduate students in Ardebil, Iran by Narimani et al. (2018), which concluded the negative effect of self-kindness and mindfulness on academic burnout.

Conclusions
The research finding supports the notion that selfcompassion has a negative effect on academic burnout, which explains that an increase in self-compassion will be followed by a reduction in academic burnout. This study suggests that undergraduate students might use self-compassion as a protective strategy to overcome difficulties in their lectures. Additionally, for undergraduate students to recover from academic burnout, support from significant others is required. Universities can develop a training program or curriculum to improve undergraduate students' selfcompassion, hence reducing academic burnout. However, this study has its limitations, one of which is the lack of diversity, since the majority of respondents came from humanities disciplines of universities in the Jabodetabek region, and none from the East Indonesia region. To generalize the research findings, researchers recommend future research to use the cluster sampling technique with geographical locations and the study major of the universities in mind. Since self-compassion can only predict academic burnout by 7.6-14.6%, future research can include other variables such as personality, self-esteem, academic motivation, perfectionism, and/or academic resiliency.