Self-Compassion and Eco-Anxiety: Examining Their Relationship in a Greek Sample

Highlights

Public health relevance—How does this work relate to a public health issue?

• Eco-anxiety is an emotional response to concerns about environmental issues.
• Environmental issues affect mental well-being at both individual and population levels and their functioning.

Public health significance—Why is this work of significance to public health?

• As awareness of environmental threats grows, people experience feelings of anxiety, stress and uncertainty about the future.
• These psychological responses can affect overall well-being, daily functioning, and quality of life.

Public health implications—What are the key implications or messages for practitioners, policy makers and/or researchers in public health?

• Different self-compassion dimensions may play distinct roles in eco-anxiety.
• Strengthen community capacity to cope with environmental change.

Abstract

Background: Eco-anxiety refers to an individual’s emotional response to environmental conditions such as air pollution and climate change. Eco-anxiety has been associated with various psychological variables; however, there is limited research on positive factors such as self-compassion. Previous studies have examined the relationship between eco-anxiety and certain dimensions of self-compassion, such as isolation and mindfulness, but have not investigated its relationship with self-compassion as a whole or with its other dimensions. Methods: The present cross-sectional study aims to address this gap by examining a sample of 586 adults of all genders from different regions of Greece. The participants were collected through convenience sampling and completed two scales: the Self-Compassion Scale and the Hogg Eco-Anxiety Scale (which has undergone a validation process in Greek). Results: Using linear multiple regression analysis, the results showed that isolation has a positive correlation with affective symptoms, rumination and behavioural symptoms. Mindfulness has a negative correlation with affective symptoms, behavioural symptoms and anxiety about personal impact. Anxiety about personal impact is positively associated with self-kindness and over-identification. Conclusions: Instead of assuming a uniform relationship between self-compassion and eco-anxiety, this dimensional approach may help clarify whether specific self-compassion processes are more relevant to particular emotional, cognitive, and behavioural manifestations of eco-anxiety.

1. Introduction

Self-compassion is defined as the kindness and understanding we show ourselves when we experience difficult situations and when we are confronted with challenging emotions and failures [1]. Although self-compassion is conceptualized as a single construct, it is also distinguished into three components with their opposites. The first dimension is self-kindness, which refers to the gentle and caring attitude an individual adopts toward themselves during difficult times. Self-kindness stands in contrast to self-judgment, where individuals respond to themselves with criticism and harshness when facing difficulties. The second dimension is common humanity, in which suffering is recognized as a shared experience and an inherent part of the human condition. Viewing pain as an inseparable aspect of human experience contrasts with isolation, where individuals perceive their struggles as unique and believe that they are the only ones who suffer. The third dimension is mindfulness, which involves awareness of one’s thoughts and feelings in the present moment. Mindfulness essentially represents a balance between avoidance, where individuals avoid engaging with negative emotions and dysfunctional thoughts, and over-identification, where they become overwhelmed by and fused with these experiences, making it difficult to separate themselves from them [1]. Self-compassion is an important resource whose cultivation promotes well-being. Showing compassion toward oneself has been associated with significant benefits, including reductions in anxiety, stress, and depression. It has also been linked to indicators of happiness, optimism, well-being, wellness, positive affect, resilience, emotional intelligence, self-regulation, and the adoption of more adaptive coping strategies for managing stress [2,3,4].

The concepts of self-compassion and mindfulness have their roots in Eastern, particularly Buddhist, contemplative traditions, where they are associated with awareness, non-judgmental attitudes, and acceptance of human experience [1]. In contemporary Western psychology, these concepts have been adapted and examined as psychological processes associated with emotion regulation and psychological well-being. Their relevance to eco-anxiety is particularly important, as the environmental crisis represents a collective, uncertain, and existential threat that may elicit intense feelings of fear, anxiety, and helplessness. From this perspective, mindfulness and self-compassion may contribute to a more adaptive way of coping with environmental distress by allowing individuals to acknowledge and accept difficult emotions without becoming excessively identified with or avoiding them.

Eco-anxiety is a relatively recent concept that has received increasing attention in the context of the ecological crisis. Despite growing interest, significant gaps remain in the literature, highlighting the need for further research [5,6]. It refers to anxiety, psychological distress, and negative emotions associated with environmental threats and risks [7,8,9], including climate change and its consequences, as well as broader environmental crises such as species extinction, ecosystem loss, pollution, and deforestation [10]. Eco-anxiety may also involve concerns about the effects of environmental threats on individuals’ lives and those of future generations [7]. Eco-anxiety is not considered a mental illness but rather a reasonable response to awareness of environmental problems and the perceived inability to address or resolve them [11]. Although it may arise directly from personal exposure to environmental problems, it more commonly reflects concerns about future environmental risks [12]. Its manifestations may include emotional symptoms such as anxiety, worry, and fear, as well as behavioral and cognitive responses, including sleep or study difficulties and rumination about the ecological crisis and the environmental consequences of one’s own behavior [10].

Eco-anxiety is not considered a mental illness but rather a reasonable response to the ecological crisis, arising from awareness of environmental problems, perceived responsibility for contributing to them, and the inability to resolve them or restore ecological balance [11]. It may develop directly following exposure to environmental problems or, more commonly, indirectly through concerns about future environmental risks [12]. Its manifestations include emotional symptoms such as anxiety, worry, and fear, as well as behavioral and cognitive responses, including sleep or study difficulties, rumination, and concerns about the environmental consequences of one’s behavior [10]. Individuals experiencing eco-anxiety may envision the future of the environment, humanity, and other living organisms as threatening and uncertain [5]. Environmental risks are perceived as threatening because of their potential consequences for both ecosystems and human social, economic, and political life, while the unprecedented scale of the ecological crisis contributes to uncertainty about the future [11]. Accordingly, eco-anxiety may involve fear, insecurity, powerlessness, lack of control, frustration, guilt, and emotional burden [5,6]. In some cases, these experiences may lead to inactivity or psychological inhibition. This may reflect either overwhelming anxiety and defensive responses such as denial and minimization, or a sense of paralysis and resignation arising from the perception that the situation is uncontrollable and unmanageable [5,12].

The relationship between self-compassion and eco-anxiety has not been extensively explored, indicating an important research gap. However, evidence from the broader literature suggests that self-compassion is negatively associated with anxiety, with individuals reporting higher self-compassion experiencing lower levels of anxiety [13,14]. Accordingly, higher self-compassion may also be associated with lower eco-anxiety, potentially because self-compassion facilitates the understanding and regulation of negative thoughts and emotions [1]. Several dimensions of self-compassion may be particularly relevant to eco-anxiety. Over-identification is associated with anxiety, as excessive focus on negative thoughts may intensify anxious experiences [1]. Conversely, self-compassion may promote acceptance of responsibility and encourage constructive action [15], which may be relevant given that eco-anxiety can be associated with feelings of paralysis and reduced engagement in preventive or corrective actions [16]. Furthermore, the dimensions of isolation and common humanity may be relevant, as eco-anxiety has been linked to loneliness, isolation, and broader existential concerns [12,17], as well as mental health difficulties such as post-traumatic stress disorder, depression, and suicidal ideation [18]. In contrast, experiences of common humanity and social connection may provide a sense of shared experience and potentially contribute to reduced eco-anxiety [1].

Mindfulness, another dimension of self-compassion, may also be relevant to eco-anxiety by promoting awareness and acceptance of thoughts and emotions without rumination, catastrophic interpretations, or avoidance [5]. Previous research suggests that the relationship between mindfulness, climate anxiety, and pro-environmental behaviour is complex and may vary across different dimensions of mindfulness and climate-related distress. Some mindfulness facets, such as acting with awareness and nonreactive decentering, have been associated with lower levels of climate anxiety, whereas greater awareness of environmental experiences may be associated with higher climate-related anxiety. Importantly, climate anxiety does not necessarily result in disengagement from environmental issues; in some cases, greater environmental distress may be associated with increased collective and personal pro-environmental action. Moreover, mindfulness may moderate the relationship between climate anxiety and pro-environmental behaviour, suggesting that its role may depend on the specific dimension of anxiety and the type of environmental response examined [19,20]. Accordingly, mindfulness-based approaches, including Cognitive Behavioral Therapy and Acceptance and Commitment Therapy, have been proposed for the management of eco-anxiety and ecological threat [21,22,23]. ACT can help individuals manage the difficult emotions elicited by environmental problems without avoiding or suppressing these emotions, while simultaneously encouraging them to act in accordance with their personal values, such as environmental protection. Core principles of ACT, including values, cognitive flexibility, adaptability, and committed action, may enhance resilience and self-care, while helping individuals cope more effectively with environmental threats and broader environmental concerns [22].

Although some evidence links self-compassion with eco-anxiety, research remains limited. In a study of 522 participants, self-compassion and acceptance of compassion from others moderated associations between connection to nature, well-being, and climate change anxiety [24]. Research has also examined self-compassion in relation to pro-environmental behavior. Studies suggest that higher self-compassion is associated with more positive environmental attitudes and greater engagement in pro-environmental behaviors [25,26,27]. Furthermore, a one-week self-compassion intervention increased both self-compassion and pro-environmental behavior while reducing materialism compared with a control group [26]. Similarly, self-compassion has been found to enhance pro-environmental behaviors and mediate the relationship between ecological identity and such behaviors [27]. Overall, these findings suggest a potential link between self-compassion and environmental concerns, while highlighting the need for further research specifically examining its relationship with eco-anxiety.

Despite the growing literature on eco-anxiety and the established relevance of self-compassion to psychological well-being and emotional regulation, limited research has directly examined the relationship between self-compassion and eco-anxiety. Moreover, little is known about whether specific dimensions of self-compassion are differentially associated with distinct manifestations of eco-anxiety. Addressing this gap, the present study examines the unique associations of the six dimensions of self-compassion—self-kindness, self-judgment, common humanity, isolation, mindfulness, and over-identification—with four dimensions of eco-anxiety: affective symptoms, rumination, behavioural symptoms, and anxiety about personal impact. By examining these constructs at the dimensional level, rather than treating self-compassion and eco-anxiety as unitary constructs, the present study aims to provide a more nuanced understanding of their relationship. Furthermore, the study contributes evidence from a Greek sample, thereby extending the emerging literature on the psychological correlates of eco-anxiety to a population that remains relatively underrepresented in this field. Therefore, the aim of the present study is to examine the association of self-compassion with eco-anxiety in the Greek population. The research question is whether self-compassion is associated with eco-anxiety. The research hypotheses are:

H1. 

The dimensions of self-compassion will be associated with affective symptoms.

H2. 

The dimensions of self-compassion will be associated with rumination.

H3. 

The dimensions of self-compassion will be associated with behavioural symptoms.

H4. 

The dimensions of self-compassion will be associated with anxiety about personal impact.

2. Materials and Methods

2.1. Sample

The sample of the study consisted of N = 586 participants, of whom the majority were women (n = 403, 68.8%), 182 were men (31.1%), and 1 participant (0.1%) did not wish to answer. Participants’ ages ranged from 19 to 69 years (M = 38.42, SD = 13.71). Most participants were married or cohabiting (n = 229, 39.1%), followed by single individuals (n = 188, 32.1%), those in a relationship (n = 113, 19.3%), divorced individuals (n = 42, 7.2%), and widowed individuals (n = 14, 2.3%). The majority of participants (n = 341, 58.2%) did not have children, followed by those with 2 children (n = 135, 23%), 1 child (n = 83, 14.2%), and 3 or more children (n = 27, 4.6%).

Regarding educational background, 219 participants (37.4%) were graduates of Bachelor’s education, 186 (31.7%) were high school graduates, 145 (24.7%) held a master’s degree, and 36 (6.1%) held a doctoral or postdoctoral degree. In terms of employment status, 403 participants (68.8%) were employed, 139 (23.7%) were students, 25 (4.3%) were unemployed, 11 (1.9%) were retired, and 8 (1.4%) were engaged in housework. Finally, 430 participants (73.4%) lived in the capital, 76 (13%) on islands, and 80 (13.6%) in other cities (

Table 1. Demographic features of the participants (Ν = 586).

		n	%	M (SD)
Gender	Man	182	31.1%
	Woman	403	68.8%
	Don’t desire to answer	1	0.1%
Marital Status	Marriage/Cohabitation	229	39.1%
	Singles	188	32.1%
	In relationship	113	19.3%
	Divorced	42	7.2%
	Widowed	14	2.3%
Children	0	341	58.2%
	1	83	14.2%
	2	135	23%
	3 and more	27	4.6%
Educational status	High school graduates	186	31.7%
	Bachelor degree	219	37.4%
	Master degree	145	24.7%
	PhD or post-PhD	36	6.1%
Employment status	Employees	403	68.8%
	University Student	139	23.7%
	Unemployed	25	4.3%
	Retired	11	1.9%
	Housework	8	1.4%
Residence	Capital city	430	73.4%
	City	80	13.6%
	Island	76	13%
Age				38.42 (13.71)

).

2.2. Materials

2.2.1. Demographic Features

At the beginning, participants completed a demographic questionnaire including questions regarding gender, age, marital status, number of children, educational status, employment status, and place of residence.

2.2.2. Hogg Eco-Anxiety Scale

To measure eco-anxiety, participants were administered the Hogg scale [10]. It includes 13 statements and consists of four factors: affective symptoms (4 items), rumination, behavioural symptoms, and anxiety about personal impact (3 items per subscale). All items are rated on a 4-point Likert scale (0 = not at all, 1 = several days, 2 = more than half the days, 3 = nearly every day). Higher scores indicate higher levels of eco-anxiety. In its original form, the scale demonstrates high reliability for all subscales (affective symptoms α = 0.92, rumination α = 0.90, behavioural symptoms α = 0.86, anxiety about personal impact α = 0.88). The scale does not produce a single total score; instead, its individual dimensions are measured separately. The Greek adaptation of the scale has undergone a validation process (the findings are not yet published) by the authors’ research team. The translation and cultural adaptation of the instrument were conducted following established procedures for cross-cultural adaptation of self-report measures. The items were translated into Greek and subsequently reviewed to ensure linguistic clarity and conceptual equivalence with the original version, using a back-translation procedure from three bilingual translators. The psychometric properties of the Greek version were also examined as part of the ongoing standardisation process in a sample of 583 participants, where the scale demonstrated satisfactory internal consistency, with a Cronbach’s alpha ranged from a = 0.826 to a = 0.869 for all of the subscales. The items followed a normal distribution, as assessed based on the mean, skewness, and kurtosis. Exploratory and confirmatory factor analyses were conducted, along with analyses of inter-item correlations and convergent and discriminant validity with related constructs, showing significant relationship (p < 0.01).

2.2.3. Self-Compassion Scale

To measure self-compassion, the Self-Compassion Scale by Neff [28] was used in a Greek validated version [29]. The scale consists of 26 items and is divided into six dimensions: the positive dimensions (self-kindness, common humanity, mindfulness) and the corresponding negative dimensions (self-judgment, isolation, overidentification). Responses are rated on a 5-point Likert scale (1 = almost never to 5 = almost always). A separate score is calculated for each subscale, as well as an overall self-compassion score.

2.3. Procedure

Participants completed the research questionnaire online after it was posted on social media (Facebook, Instagram) and were asked to share the survey with other potential participants. The recruitment post informed them about the purpose of the study and the inclusion criteria (only individuals aged 18 years or older, with a good command of the Greek language and residing in Greece, were eligible to participate in the study). They were then informed through an information sheet about their rights and subsequently provided informed consent. The questionnaire took approximately 10 min to complete and the study was conducted from January to April 2026. A total of 603 individuals completed the questionnaire, of whom 9 did not complete the questionnaire and 8 did not meet the eligibility criteria. Therefore, the final sample consisted of N = 586 participants. The study complied with ethical and research guidelines, ensuring anonymity, confidentiality, full transparency regarding the purpose of the study, and participants’ right to withdraw at any time. The study was approved by the Ethics and Deontology Committee of Panteion University of Social and Political Sciences (protocol number 3/27-01-2026).

3. Results

3.1. Descriptive Statistics

Initially, descriptive statistics were conducted in order to calculate the means and standard deviations of the study variables, followed by reliability analysis. The variables were found to be at moderate levels on the self-compassion scale: self-kindness (M = 3.12, SD = 0.87), common humanity (M = 3.16, SD = 0.88), mindfulness (M = 3.27, SD = 0.88), self-judgment (M = 2.94, SD = 0.94), isolation (M = 2.74, SD = 0.85), and over-identification (M = 3.03, SD = 0.98). Moderate levels were also found for the dimensions of eco-anxiety: affective symptoms (M = 0.98, SD = 0.75), rumination (M = 0.85, SD = 0.72), behavioural symptoms (M = 0.51, SD = 0.29), and personal impact (M = 0.96, SD = 0.73). Reliability was high for all variables: self-compassion (α = 0.906), self-kindness (α = 0.833), common humanity (α = 0.771), mindfulness (α = 0.822), self-judgment (α = 0.841), isolation (α = 0.816), over-identification (α = 0.813), affective symptoms (α = 0.871), rumination (α = 0.870), behavioural symptoms (α = 0.838), and personal impact (α = 0.829). Finally, a normality test of the distributions was conducted, which showed that the variables met the assumptions for parametric analyses (p > 0.05) (

Table 2. Descriptive statistics (N = 586).

	M	SD	Cronbach’s Alpha	Kolmogorov–Smirnov
Self-kindness	3.12	0.87	0.833	0.065
Common Humanity	3.16	0.88	0.771	0.068
Mindfulness	3.27	0.88	0.822	0.092
Self-judgment	2.94	0.94	0.841	0.070
Isolation	2.74	0.85	0.816	0.078
Over-identification	3.03	0.98	0.813	0.088
Affective symptoms	0.98	0.75	0.871	0.111
Rumination	0.85	0.72	0.870	0.153
Behavioural symptoms	0.51	0.29	0.838	0.286
Anxiety about personal impact	0.96	0.73	0.829	0.145

).

3.2. Correlations Among Variables

Pearson correlations among the study variables are presented in

Table 3. Associations among self-compassion and eco-anxiety dimensions (N = 586).

Variables	1	2	3	4	5	6	7	8	9	10
(1) Self-kindness	1
(2) Self-judgment	−0.252 **	1
(3) Common humanity	0.673 **	−0.036	1
(4) Isolation	−0.260 **	0.714 **	−0.121 **	1
(5) Mindfulness	0.708 **	−0.064	0.689 **	−0.190 **	1
(6) Overidentification	−0.182 **	0.726 **	−0.076	0.738 **	−0.185 **	1
(7) Affective symptoms	−0.080	0.166 **	−0.020	0.245 **	−0.127 **	0.220 **	1
(8) Rumination	0.015	0.132 **	0.060	0.191 **	−0.024	0.148 **	0.683 **	1
(9) Behavioural symptoms	−0.094 *	0.184 **	−0.037	0.268 **	−0.112 **	0.172 **	0.599 **	0.370 **	1
(10) Anxiety about personal impact	0.039	0.219 **	0.057	0.220 **	−0.045	0.260 **	0.573 **	0.651 **	0.335 **	1

* p < 0.05, ** p < 0.01.

. The self-compassion dimensions showed differential patterns of association with the dimensions of eco-anxiety, with some dimensions showing significant associations whereas others were not significantly associated. These bivariate associations provide an overview of the relationships among the study variables prior to the multiple regression analyses.

3.3. Multiple Linear Regression

A multiple linear regression was conducted to examine whether the dimensions of self-compassion are associated with the dimensions of eco-anxiety. In the first model, the dimensions of self-compassion were entered as predictor variables, with affective symptoms as the outcome variable. The overall model showed that the dimensions of self-compassion explained 7.8% of the variance in affective symptoms. The overall regression model was statistically significant, F(6, 579) = 8.156, p < 0.001, R² = 0.078, adjusted R² = 0.068. Isolation was significantly positively associated with affective symptoms, B = 0.148, SE = 0.048, β = 0.198, t(579) = 3.068, 95% CI [0.053, 0.243], p = 0.002. Mindfulness was significantly negatively associated with affective symptoms, B = −0.142, SE = 0.055, β = −0.168, t(579) = −2.594, 95% CI [−0.249, −0.034], p = 0.010. The other dimensions (self-kindness, self-judgment, common humanity, and overidentification) were not significantly associated with affective symptoms (p > 0.05) (

Table 4. Multiple regression analysis predicting affective symptoms from self-compassion dimensions (N = 586).

Independent Variables	B	SE B	β	t	95% CI	p
Constant	0.587	0.178		0.303	[0.238, 0.935]	0.001
Self-kindness	0.019	0.056	0.022	0.341	[−0.091, 0.130]	0.733
Self-judgment	−0.029	0.053	0.036	0.537	[−0.133, 0.076]	0.591
Common humanity	0.093	0.051	0.110	1.840	[−0.006, 0.192]	0.066
Isolation	0.148	0.048	0.198	3.068	[0.053, 0.243]	0.002
Mindfulness	−0.142	0.055	−0.168	−2.594	[−0.249, −0.034]	0.010
Over-identification	0.062	0.051	0.081	1.205	[−0.039, 0.163]	0.229

R² = 0.078, adjusted R² = 0.068, F(6, 579) = 8.156, p < 0.001. B = Unstandardized B, SE = Coefficients Std. Error, β = Standardized Coefficient.

).

Regarding the second dimension of eco-anxiety, rumination, the model explains the 4.9% of its variance. The overall model was statistically significant F(6, 579) = 4.993, p < 0.001, R² = 0.049, adjusted R² = 0.039. Only isolation was positively associated with rumination, B = 0.144, SE = 0.048, β = 0.198, t(579) = 3.016, 95% CI [0.050, 0.237], p = 0.003. The other dimensions were not statistically significant (p > 0.05) (

Table 5. Multiple regression analysis predicting rumination from self-compassion dimensions (N = 586).

Independent Variables	B	SE B	β	t	95% CI	p
Constant	0.274	0.175		1.564	[−0.070, 0.619]	0.118
Self-kindness	0.062	0.056	0.074	1.112	[−0.047, 0.171]	0.267
Self-judgment	0.007	0.053	0.009	0.131	[−0.096, 0.110]	0.896
Common humanity	−0.096	0.050	0.116	1.918	[−0.002, 0.194]	0.056
Isolation	0.144	0.048	0.198	3.016	[0.050, 0.237]	0.003
Mindfulness	−0.098	0.054	−0.119	−1.811	[−0.204, 0.008]	0.071
Over-identification	−0.003	0.051	−0.004	−0.065	[−0.103, 0.096]	0.948

R² = 0.049, adjusted R² = 0.039, F(6, 579) = 4.993, p < 0.001. B = Unstandardized B, SE = Coefficients Std. Error, β = Standardized Coefficient.

).

Regarding the behavioural symptoms, the model explains the 8.1% of the variance in behavioural symptoms. The overall model was statistically significant, F(6, 579) = 8.513, p < 0.001, R² = 0.081, adjusted R² = 0.072. Isolation was positively associated with behavioural symptoms, B = 0.204, SE = 0.044, β = 0.296, t(579) = 4.581, 95% CI [0.116, 0.291], p < 0.001. Mindfulness was negatively associated with behavioural symptoms, B = −0.106, SE = 0.050, β = −0.136, t(579) = −2.103, 95% CI [−0.205,−0.007], p = 0.036. The other variables were not significant (p > 0.05) (

Table 6. Multiple regression analysis predicting behavioural symptoms from self-compassion dimensions (N = 586).

Independent Variables	B	SE B	β	t	95% CI	p
Constant	0.173	0.164		1.058	[−0.148, 0.494]	0.291
Self-kindness	0.020	0.052	0.025	0.384	[−0.082, 0.122]	0.701
Self-judgment	0.028	0.049	0.038	0.563	[−0.069, 0.124]	0.574
Common humanity	0.054	0.047	0.069	1.168	[−0.037, 0.146]	0.243
Isolation	0.204	0.044	0.296	4.581	[0.116, 0.291]	<0.001
Mindfulness	−0.106	0.050	−0.136	−2.103	[−0.205, −0.007]	0.036
Over-identification	−0.063	0.047	−0.089	−1.328	[−0.156, 0.030]	0.185

R² = 0.081, adjusted R² = 0.072, F(6, 579) = 8.513, p < 0.001. B = Unstandardized B, SE = Coefficients Std. Error, β = Standardized Coefficient.

).

Finally, the overall model explains 9.5% of the variance in anxiety about personal impact and was statistically significant, F(6, 579) = 10.089, p < 0.001, R² = 0.095, adjusted R² = 0.085. Self-kindness was positively associated with anxiety about personal impact, B = 0.157, SE = 0.054, β = 0.186, t(579) = 2.874, 95% CI [0.050, 0.263], p = 0.004, as well as overidentification (B = 0.104, SE = 0.050, β = 0.140, t(579) = 2.095, 95% CI [0.007, 0.202], p = 0.037). Mindfulness was negatively associated (B = −0.157, SE = 0.053, β = −0.190, t(579) = −2.964, 95% CI [−0.261, −0.053], p = 0.003). The other dimensions were not significant (p > 0.05) (

Table 7. Multiple regression analysis predicting anxiety about personal impact from self-compassion dimensions (N = 586).

Independent Variables	B	SE B	β	t	95% CI	p
Constant	0.078	0.172		0.454	[−0.259, 0.416]	0.650
Self-kindness	0.157	0.054	0.186	2.874	[0.050, 0.263]	0.004
Self-judgment	0.088	0.052	0.114	1.712	[−0.013, 0.189]	0.087
Common humanity	0.070	0.049	0.084	1.429	[−0.026, 0.166]	0.154
Isolation	0.042	0.047	0.058	0.908	[−0.049, 0.134]	0.364
Mindfulness	−0.157	0.053	−0.190	−2.964	[−0.261, −0.053]	0.003
Over-identification	0.104	0.050	0.140	2.095	[0.007, 0.202]	0.037

R² = 0.095, adjusted R² = 0.085, F(6, 579) = 10.089, p < 0.001. B = Unstandardized B, SE = Coefficients Std. Error, β = Standardized Coefficient.

).

4. Discussion

The purpose of the current study was to test whether self-compassion is associated with the dimensions of eco-anxiety. The findings from the multiple linear regression showed that many of the dimensions of self-compassion are linked with the dimension of eco-anxiety. In summary, isolation has a positive correlation with affective symptoms, rumination and behavioural symptoms. Mindfulness has a negative correlation with affective symptoms, behavioural symptoms and anxiety about personal impact, whether anxiety about personal impact is associated positively with self-kindness and over-identification. These different findings highlight the need for further consideration of what is ultimately occurring in the relationship between self-compassion and eco-anxiety.

The first hypothesis proposed that self-compassion would be associated with the affective symptoms of eco-anxiety. The results indicated that the dimensions of self-compassion, considered collectively, were significantly associated with affective symptoms. However, when examining the unique association of each dimension, isolation was found to be positively associated with affective symptoms, whereas mindfulness was negatively associated with affective symptoms. Isolation represents one of the negative dimensions of self-compassion, and the findings suggest that higher levels of isolation were associated with higher levels of affective symptoms. One possible explanation is that feelings of isolation may increase vulnerability to environmental threats. When individuals feel alone or perceive a lack of adequate social support, they may experience greater difficulty regulating emotions such as fear, worry, sadness, or guilt associated with climate change. Consequently, social isolation may be associated with more intense emotional responses to the environmental crisis. Consistent with previous research, individuals experiencing eco-anxiety may also experience loneliness and isolation due to the existential nature of the issues associated with eco-anxiety [12,17]. In contrast, mindfulness was negatively associated with affective symptoms, indicating that higher levels of mindfulness were associated with lower levels of affective symptoms. Previous research has suggested that mindfulness is associated with lower levels of eco-anxiety, potentially through greater engagement in prosocial environmental behaviours [5,19,20]. Furthermore, mindfulness-based interventions have been examined as a potential approach for reducing eco-anxiety [21,22,23]. Although the overall regression model was statistically significant, the remaining dimensions of self-compassion did not make a significant unique association when all dimensions were considered simultaneously. Previous research has demonstrated that self-compassion is generally associated with lower levels of anxiety [13,14]; however, research specifically examining the relationship between self-compassion and eco-anxiety remains limited. Nevertheless, related research has indicated that self-compassion is positively associated with prosocial behaviour [25,26,27] and may moderate the relationship between connection to nature and well-being, with stronger connections to nature being associated with lower climate change anxiety [24]. In the present study, however, not all dimensions of self-compassion demonstrated a statistically significant unique association to affective symptoms. It was initially expected that self-kindness would be negatively associated with affective symptoms, as self-kindness may facilitate the regulation and management of negative emotions [1]. However, the ability to regulate negative emotions does not necessarily prevent their occurrence. Thus, individuals who are more self-compassionate may be better able to cope with negative emotions elicited by environmental threats, without necessarily experiencing fewer such emotions. Similarly, self-judgment was not significantly associated with affective symptoms. As with self-kindness, this may be partly explained by the nature of eco-anxiety as a response to a broader and global environmental threat rather than to a perceived personal failure. Consequently, self-judgment may be less directly relevant to the emotional responses associated with climate-related concerns. Finally, neither common humanity nor over-identification demonstrated a statistically significant association with affective symptoms. A negative association between common humanity and affective symptoms might have been expected, as individuals who perceive themselves as part of a broader human community may experience greater social connectedness and feel less alone in confronting environmental concerns. However, this sense of shared experience may not necessarily prevent the emotional responses associated with eco-anxiety. Conversely, over-identification might have been expected to be positively associated with affective symptoms, given that becoming excessively immersed in one’s thoughts and emotions may intensify psychological distress. Nevertheless, over-identification may be more closely related to the cognitive and ruminative aspects of eco-anxiety than to the emotional symptoms captured by the affective symptoms dimension. Overall, these findings suggest that the relationship between self-compassion and eco-anxiety may be complex and may differ according to the specific dimension of eco-anxiety being examined.

Moving on to the second research hypothesis, the dimensions of self-compassion were expected to be associated with rumination. The findings indicated that only isolation has a statistically significant association with rumination, where those with high scores on isolation reported higher levels of rumination. Feelings of isolation are prevalent among individuals experiencing eco-anxiety, potentially due to the existential nature of environmental threats [12,17]. Consequently, individuals who experience greater social disconnection may be more likely to engage in persistent rumination regarding environmental issues. Nevertheless, self-kindness, self-judgment, common humanity, mindfulness, and over-identification did not demonstrate statistically significant associations with rumination. One possible explanation for these non-significant associations is that rumination may reflect a more specific cognitive response to climate-related concerns rather than a general tendency in the way individuals relate to themselves. These dimensions of self-compassion may influence how individuals respond to or regulate distressing thoughts without necessarily reducing the frequency or persistence of climate-related thoughts. Furthermore, the non-significant findings may suggest that other factors, such as perceived threat, uncertainty about the future, environmental concern, or perceived lack of control, may be more directly relevant to climate-related rumination. Overall, these findings suggest that social connectedness may be particularly relevant to the cognitive dimension of eco-anxiety, whereas other dimensions of self-compassion may have more limited or indirect associations with rumination.

The third research hypothesis examined the potential relationship between the dimensions of self-compassion and behavioural symptoms. Although the overall regression model was statistically significant, when the unique association of each dimension of self-compassion was examined, the findings were consistent with those of the first research hypothesis: only isolation and mindfulness were associated with behavioural symptoms. Isolation was positively associated with behavioural symptoms, suggesting that individuals who experience greater feelings of isolation may be more likely to report behavioural symptoms related to eco-anxiety. One possible explanation is that a lack of social connection and support may make it more difficult for individuals to regulate their responses to environmental concerns, potentially leading to more pronounced behavioural reactions [12,17]. In contrast, mindfulness was negatively associated with behavioural symptoms, suggesting that higher levels of mindfulness may be associated with fewer behavioural symptoms. This may be because mindfulness can facilitate greater awareness and acceptance of one’s thoughts and emotions, allowing individuals to respond to environmental concerns in a more regulated manner rather than reacting behaviourally to distress [5,19,20,21,22,23]. One possible explanation for the non-significant associations between the remaining dimensions of self-compassion and behavioural symptoms is that these dimensions may not make a unique association with behavioural responses to eco-anxiety when considered simultaneously with isolation and mindfulness. While self-kindness, self-judgment, common humanity, and over-identification may influence how individuals experience or regulate emotional distress, behavioural responses to climate-related concerns may depend more strongly on social connectedness and the way individuals attend to and respond to their internal experiences. Additionally, the overlap between the dimensions of self-compassion may have reduced their unique role in the regression model. Therefore, the non-significant findings should not be interpreted as evidence that these dimensions are unrelated to eco-anxiety, but rather that they did not make a statistically significant unique association with behavioural symptoms in the present model.

The final research hypothesis examined whether the dimensions of self-compassion are associated with anxiety about personal impact. The results indicated that the overall regression model is significantly linked to anxiety about personal impact; however, when the unique association of each dimension was examined, only self-kindness, mindfulness, and over-identification were associated. Mindfulness was negatively associated with anxiety about personal impact, whereas over-identification was positively associated with this dimension of eco-anxiety. These findings are consistent with the theoretical framework of self-compassion [1], according to which mindfulness involves awareness of one’s thoughts and emotions in the present moment, potentially reducing excessive rumination and over-identification. Mindfulness may therefore be associated with lower levels of anxiety regarding one’s personal environmental impact. Previous research examining the relationship between mindfulness and eco-anxiety [5,19,20,21,22,23] provides further support for this interpretation. Nevertheless, the positive association between self-kindness and anxiety about personal impact was noteworthy, as self-kindness might initially be expected to function as a protective factor. One possible explanation for this unexpected finding is that individuals who demonstrate greater self-kindness may also be more sensitive to environmental concerns and more aware of their personal responsibility and role in addressing environmental issues. Thus, self-kindness may coexist with greater awareness of one’s environmental impact without necessarily reducing concerns about it. The remaining dimensions of self-compassion did not make a statistically significant unique association with anxiety about personal impact. One possible explanation is that a sense of belonging or isolation may not be directly related to an individual’s perception of personal responsibility towards environmental problems. Similarly, self-judgment may not represent one of the primary mechanisms underlying eco-anxiety, as individuals may perceive their personal association to broader global environmental threats as relatively limited. Consequently, concerns regarding personal environmental impact may be more closely associated with mindfulness, self-kindness, and over-identification than with social connectedness or self-evaluative processes.

Nevertheless, it is important to note that the overall model, including the dimensions of self-compassion, was significantly associated with all dimensions of eco-anxiety. Since eco-anxiety involves feelings of environmental threat [11], individuals with higher levels of self-compassion may be better able to manage this sense of threat. Self-compassion may also facilitate better adaptation to difficult situations [15]; therefore, individuals with greater self-compassion may adapt more effectively to circumstances involving environmental threats or feel more secure when facing such challenges. On the other hand, individuals experiencing high levels of eco-anxiety may show limited engagement in action and may not engage in preventive or corrective behaviours, which can, in turn, intensify eco-anxiety [16]. In contrast, self-compassion has been associated with taking action rather than passively accepting difficult circumstances [1], which may help explain the significant findings observed for anxiety about personal impact.

However, this association appears to vary across the different dimensions of eco-anxiety and the individual dimensions of self-compassion. The differential pattern of findings may suggest that the dimensions of self-compassion are associated with distinct facets of eco-anxiety, with some dimensions showing more consistent associations across outcomes (e.g., isolation and mindfulness). Nevertheless, as this is one of the first studies to examine the relationship between self-compassion and eco-anxiety, and given the cross-sectional nature of the study and its focus on a Greek sample, the findings should be interpreted with caution. In particular, the cross-sectional design does not allow causal inferences to be drawn regarding the direction of these associations. Future research using longitudinal designs and more diverse samples would be valuable for clarifying the nature and direction of the relationship between self-compassion and the different dimensions of eco-anxiety.

4.1. Limitations

At this point, it is necessary to acknowledge the limitations of the study, such as its cross-sectional design, which does not allow for generalization of the results over time. In addition, the sample was collected through convenience sampling, where the non-probability nature of the sampling method does not allow the generalization of the findings to the representative population. Although the sample size was large, there was overrepresentation in certain categories, such as women, employed individuals, and residents of the capital. Furthermore, the sample covered a wide age range (19 to 69 years), spanning different generations (e.g., Gen Z, Boomers), which may reflect different attitudes toward environmental issues. The cross-sectional design does not allow causal or directional conclusions to be drawn. Second, all variables were assessed using self-report measures, which may be subject to reporting biases and may not fully reflect participants’ actual experiences or behaviours, raising the possibility of common method variance. Furthermore, social desirability may have influenced responses, particularly given the socially and morally salient nature of environmental concerns. Another limitation is that demographic covariates were not included in the regression models. Although the sample showed an unequal gender distribution, there was insufficient prior evidence to establish which demographic characteristics should be considered potential confounders in the relationship between self-compassion and eco-anxiety. As this study represents an initial investigation of this relatively understudied association, future research should examine whether demographic and other contextual factors influence or account for these associations. Finally, although the Greek adaptation of the questionnaire has undergone a validation process, the findings have not yet been published in a peer-reviewed journal. Therefore, the psychometric properties of the Greek version should be considered when interpreting the present findings.

4.2. Future Research

Future research could take the above limitations into account when conducting subsequent studies. Future research should further investigate the role of sociodemographic characteristics in eco-anxiety. Ongoing research is currently examining the sociodemographic profile of eco-anxiety in greater depth, which may contribute to a more comprehensive understanding of individual differences in eco-anxiety and the factors associated with its expression. Although the findings do not provide direct evidence for the effectiveness of interventions, they may provide preliminary insights into potential psychological factors that could be considered in future research on interventions targeting eco-anxiety.

5. Conclusions

Importantly, the present findings should not be interpreted as suggesting that lower levels of eco-anxiety are necessarily indicative of better psychological adaptation. Eco-anxiety may also represent an understandable and potentially adaptive response to genuine environmental threats and may motivate individuals to engage in pro-environmental action. From this perspective, self-compassion may be more relevant to the regulation of the distress associated with eco-anxiety than to its elimination. By allowing individuals to acknowledge difficult emotions without becoming overwhelmed or excessively identified with them, self-compassion may potentially facilitate a more balanced response to environmental concerns while preserving motivation for constructive environmental engagement. Neff [1] argued that self-compassion creates a more compassionate world. Since eco-anxiety is a collective rather than an individual experience, different self-compassion dimensions may play distinct roles in eco-anxiety. The present study may contribute to a more nuanced theoretical understanding of eco-anxiety by examining whether different dimensions of self-compassion are differentially associated with distinct manifestations of eco-anxiety. Rather than assuming a uniform relationship between self-compassion and eco-anxiety, this dimensional approach may help clarify whether specific self-compassion processes are more relevant to particular emotional, cognitive, and behavioural manifestations of eco-anxiety.