Nourishing Compassion in Finnish Kindergarten Head Teachers: How Character Strength Training Influences Teachers’ Other-Oriented Behavior

Compassion refers to noticing, feeling and acting to alleviate suffering in others. Being one of the crucial socio-emotional and cultural skills, surprisingly little is known of early childhood education (ECE) teachers’ compassionate behavior and their ability to teach it to children. Overall, research on compassion in early childhood education and care (ECEC) settings is scarce. However, existing studies show that compassion skills can be acquired through training, and a collective compassion capability can be cultivated by everyday practices. As part of the positive psychology movement, compassion can be viewed as one of the character strengths (Peterson and Seligman 2004), generic skills, or “soft skills” that are underlined in the latest curricula and pedagogical documents (OECD 2017). In our present study, 95 ECE head teachers participated in the interventions, the aim of which was promoting compassion, character strengths and a supportive organization culture. A classic controlled pre/post-measuring protocol was applied to monitor progress. Additionally, qualitative material was collected from 33 participants. The intervention participants showed a significant increase in their identification and usage of strengths and in creating a supportive organizational culture. Content analysis of the qualitative statements revealed three main themes, Caring professionalism, Warm presence and Positive feedback, evidencing strivings toward an emotionally and professionally supportive ECEC organization.


Introduction
The purpose of the present study was to investigate the outcomes of a compassion training program for early childhood education (ECE) head teachers in Finnish kindergartens. Compassion (Latin "cum patior") literally means "with suffering" (e.g., Maben et al. 2010, p. 9;Strauss et al. 2016). The concept of compassion is complex, and several interpretations have emerged from a range of traditions. In all of them, certain characteristics are involved, namely noticing the pain in another person and having a willingness to help them (Lazarus and Lazarus 1991, p. 289;Strauss et al. 2016). The ability to experience or share the pain of another person and to form warm and understanding feelings toward that person to ease their suffering is a ground-laying skill in human beings (Nussbaum 1996).
Philosophical, theological, evolutionary and psychological approaches have been applied in the study of human compassionate behavior starting from the ancient Greek philosophers. According to Aristotle (2009), one can feel compassion when serious, undeserved suffering in another person becomes visible and one can relate to the sufferer themselves.
In the greatest Abrahamic religions, Judaism, Christianity and Islam, compassion is seen as a way for humans to emulate God's love. In the Biblical tradition, compassion is "a duty to divine law" (Wuthnow 1991, p. 50). In the Qu'ran (30:21) it is said "He put between you love and compassion". In Buddhist philosophy, compassion is a central theme, "the basic nature of human beings". The Dalai Lama (1995) has stated that "genuine compassion must have both wisdom and loving-kindness".
From an evolutionary point of view, compassion seems to be reproductively advantageous since it helps in nurturing and taking care of offspring (Gilbert 2009). Twin studies on human social behavior genetics have provided evidence that the capacity to feel empathy and have the potential to act compassionately is at least partly genetically determined (Chakrabarti et al. 2009;Ebstein et al. 2010). This heritage has guaranteed social bonding and human survival throughout history (Gilbert 2009).
Professor Jane Dutton is a pioneer in compassion research in organizational positive psychology. According to her, the process of compassion can be divided into phases of noticing the pain in another person, feeling empathic concern, and acting in order to alleviate suffering in others (Dutton et al. 2014;Kanov et al. 2004). The three phases of compassion do not necessarily follow each other in the given order but overlap and flow back and forth in the process. Being able to act in a compassionate way enables a plenitude of other-directed emotions (e.g., empathy, kindness and forgiveness), creating chances of better mutual understanding in an organization (Dutton et al. 2007). Feeling cared for and supported by others improves employees' loyalty, dedication, and engagement in the workplace. Furthermore, people are more willing to cooperate and help each other in organizations in which compassion prevails (Fineman 2000).
Professor Dutton has stated that compassion is both timeless and timely (Dutton et al. 2014). It is timeless due to its ubiquitous nature and centrality in religious and philosophical writings over the centuries. The timeliness of compassion arises from the growing understanding of the importance of interpersonal dynamics in organizations, communities, and every place where people interact (Dutton et al. 2014). Being kind, empathic, and "nice" are valued trends in today's economy and labor market (e.g., Friedman and Gerstein 2017). Being willing to understand and appreciate other people is also at the heart of positive psychology. "Other people matter", is the widely-cited sentence by Professor Christopher Peterson, one of the founders of the positive psychology movement (see Seligman 2012, p. 20).
According to Vietnamese Zen monk Hanh (2006), compassion is a verb. In that definition, the difference from empathy becomes clear. When empathy suggests the idea of stepping into somebody's shoes, compassion goes further. Compassion contains action, doing good deeds to the benefit of someone else.
Altruism is another concept close to compassion. An altruistic person acts to help others (not closely related) without thinking of their own good. However, unlike in a compassionate act, the receiver of help does not have to be troubled. Rather, altruistic actions may be directed to anyone. Furthermore, an altruistic person may be at risk of danger herself while offering help (Fehr and Fischbacher 2003). In compassion, no similar sacrifices are necessarily involved, but a compassionate person may experience satisfaction while benefiting the other one (Atkins and Parker 2012).
In our current study, we adopted the definition of Dutton et al. (2014) of compassion as a noticing-feeling-acting process. We further emphasize the role of kindness in the process of compassion. Compassion and kindness are related, but not synonymous constructs. Neff, for example, describes kindness as one part of self-compassion (see e.g., Neff et al. 2007). However, compassion can exist without kindness. For example, Strauss et al. write (2016, p. 19) that when giving someone "tough love" (that is, when we are hard in order to help a person in the long run), we are being compassionate, but not necessarily kind. On the other hand, kindness does not require noticing suffering or any emotions in another person. For instance, we can be kind and let someone pass us in a line without even seeing the person properly. In our definition, kindness paves the way for compassion. It is the first step out of solipsism and toward other people (Peterson and Seligman 2004, p.326).
In organizations such as kindergartens, the way adults respond to emotions expressed by each other is a living demonstration for children of how to build caring interactions with other people Taggart 2016). Many studies have shown that socio-emotional learning occurs at an early age. With positive adult role models to follow, children as young as one year of age, learn prosocial behavior, defined as "voluntary behavior intended to benefit another" (Eisenberg et al. 2006, p. 646).
Early childhood educators are often the first non-family adults who regularly interact with children (Smith 2013). Thus, showing compassion in kindergarten is of great importance in supporting children's other-directed behavior. Compassionate behavior in early childhood helps in creating friendships (Wang et al. 2019) and it predicts closer relationships, lower rates of bullying, and higher self-confidence in later childhood (Ashiabi 2007;Eisenberg et al. 2006). Also, it prepares children for school, how to adapt in a student group, and how to manage academic tasks (Walker et al. 2004).
Furthermore, compassionate adults working with children not only model caring behavior, but they are able to create a better-functioning organizations in which all members feel included. In such organizations, psychological safety is improved, positive emotions, such as gratitude, are evoked, anxiety is reduced (Lilius et al. 2008, see also Seppala et al. 2013) and the sufferer's attachment and commitment to the organization are improved (Grant andMayer 2009, see also, Dutton et al. 2014;Rajala et al. 2019). In addition, the person who is acting compassionately can feel the sense of satisfaction that arises from helping others. This may help in considering oneself to be a caring person (Grant and Mayer 2009). These are all factors that create a supportive work culture, strengthen mutual trust, enhance work-related well-being and, again, contribute to compassion.

Compassion in Positive Psychology Research
In our current study, the theory base comes from positive psychology and the seminal work by Peterson and Seligman (The Handbook of Character Strengths and Virtues 2004) at the beginning of the new millennium (see also Csikszentmihalyi and Seligman 2000). One of the main topics in positive psychology is the study of positive individual traits and that which constitutes good character. In their groundbreaking research, Peterson and Seligman (2004) classified six core virtues: wisdom, courage, humanity, justice, temperance, and transcendence, that represent morality and "what is good of a person". They further identified 24 character strengths that are distinguishable routes to displaying one or another of the six core virtues (Peterson and Seligman 2004, pp. 10-13). According to the criteria set by Peterson and Seligman (2004, pp. 16-28), character strengths contribute to fulfilment that constitutes a good life, are morally valued in their own right, and can be manifested in the range of an individual's behavior. Since the publication of the research by Peterson's and Seligman in 2004, hundreds of studies have been conducted to learn, teach and assess character strengths (see viacharacter.org/research for a comprehensive list of the literature). Also, in schools, character strength interventions are numerous (for a review, see Lavy 2019).
The character strengths of kindness, love, and social intelligence make up the virtue of humanity. According to Peterson and Seligman (2004, p. 326), kindness, generosity, nurturance, care, compassion, and altruistic love are a network of closely related terms. The use of these "strengths of heart" predicts the ability to build and maintain social relationships (Wagner et al. 2019). Living a compassionate life seems to facilitate social bridging and by that means, adds to one's happiness (e.g., Fredrickson 2013). This notion is critical since it contradicts the lay belief that compassion, conceptualized as "having a heavy load of someone else's suffering on your shoulders" would only lead to unhappiness (Figley 1995).
As written by Peterson and Seligman (2004, p. 31), their classification of the 24 character strengths is not a "finished product". Specific strengths might be added, deleted, or combined as theory and research proceed. Accordingly, in the current study and in all our research since 2015, we have defined compassion as a character strength in its own right, not just a constituent part of kindness Vuorinen et al. 2019, in press).

Teaching Compassion and Other Character Strengths in Early Childhood Education
Based on our literature review, surprisingly little is known about early childhood education (ECE) teachers' own compassionate behavior and their ability to teach it to children (see also Lipponen et al. 2018). More is known about how ECE teachers demonstrate related character strengths such as love, kindness and forgiveness (Haslip et al. 2019). As written by Haslip et al. (2019, p. 532), for children to develop values and character strengths, adults need to model and demonstrate them (see also, Berkowitz and Bier 2014;Kokoszka and Smith 2016). ECE teachers regularly practice socialization with young children and are competent at engaging with them, which in turn can nurture character strengths (FitzSimons 2015). Furthermore, Taggart (2016, p. 179) points out the core responsibilities of ECE teachers: The professional mandate of early years practitioners is to water the seed of human flourishing itself and foster compassion in future society.
In Finland, Lipponen et al. (2018;Hilppö et al. 2019;Rajala et al. 2019) have studied compassion from a sociocultural perspective, emphasizing the importance of creating cultures of compassion rather than promoting compassionate individual practitioners in ECEC settings. Their studies highlight the centrality of socially shared practical wisdom in ECEC settings when building a compassionate educational environment. By shared practical wisdom Lipponen et al. mean collaboration between ECEC workers and children in nurturing a compassionate kindergarten culture ). According to them, ECE teachers' own skills to model and explicitly teach how to practice kindness and compassion is a priority in developing socio-emotional skills in children and in creating a more caring preschool culture. The ability to live with other people, with the willingness to understand and relate to their feelings, constitutes the foundation of social relationships. Managing feelings and learning to cope with a variety of people with a compassionate mindset can be promoted among preschool-aged children in ECEC settings (Shoshani and Slone 2017). With competent educators, the path to grow to a positively other-oriented adulthood is given an optimal start.
As part of the positive psychology movement, compassion is one of the character strengths, generic skills, or "soft skills" underlined in the latest curricula and pedagogical documents (OECD 2017). Compassion, like other character strengths, should be taught early and explicitly (Spinrad and Eisenberg 2014). Although the ability to have other-oriented emotions is in part genetically determined, compassion is a skill that grows with training; a skill which children can learn to both express and receive (Gilbert 2009). Character strength interventions have been conducted in various 1 3 settings (e.g., Niemiec 2017, for a list of studies, see viacharacter.org/research). Although compassion training programs have had proven success in clinical populations (Gilbert and Procter 2006;Kirby 2017), compassion is not among the most often trained skills in character strength interventions. Rather, it has been introduced as part of other strengths of humanity (for example, see Güsewell and Ruch 2012).
In most cases, compassion training interventions have been based on practicing meditation (e.g., Pace et al. 2009) and loving-kindness meditation (e.g., Seppala et al. 2014). Explicit compassion training programs are still few. In the Center for Compassion and Altruism Research and Education (CCARE) at Stanford University School of Medicine, studies on compassion and altruistic behavior and how to promote their uses have shown promising results. The outcomes manifest include increases in positive emotions, and decreases in illness symptoms, depression, and self-criticism (see https ://ccare .stanf ord.edu/resea rch/peer-revie wed-ccare -artic les/ for a list of publications). However, as far as we know, the participants of the studies have not been ECEC professionals.
Our current study is one of the few non-meditation-based programs aimed at explicitly teaching what compassion and kindness mean, and how they can be nourished in an ECEC setting (see also Uusitalo and Vuorinen 2020). We wish to fill the gap in positive psychology research by studying how kindergarten head teachers' compassionate behavior can be improved in order to benefit the whole organization.

Research Questions
The purpose of the current study was to promote a positive, other-oriented, and supportive work culture in ECEC settings by providing training in compassion and other character skills to ECE head teachers via a two-month intervention. Based on previous research, we hypothesized that a compassion intervention would lead to improvements in several dimensions of well-being in comparison to the control group. Specifically, we expected the compassion training to result in significantly higher increases in the head teachers' sense of compassion, identification of character strengths, usage of character strengths, ability to create a supportive work climate, and ability to find meaning in work, when compared to a matched control group. Also, we wanted to know what other-oriented kind acts the head teachers do in their daily work.
Our research questions were: (1) Does attendance in early childhood compassion training affect the participants' sense of compassion? (2) Is the participants' ability to identify character strengths improved after the training?
(3) Is the participants' ability to use character strengths improved after the training? (4) Does attendance in training affect the participants' ability to create a supportive work climate? (5) Are the participants more able to find meaning in work after the training? (6) What other-oriented kind acts do the participants do in their daily work?

Background to the Present Study
Our current study is part of a large, multidisciplinary project called CoPassion (see copassion.fi) aimed at fostering compassionate behavior in workplaces. In the CoPassion project, three methods were exploited: teaching compassion and other emotion skills (Paakkanen et al. 2019), teaching self-compassion (Ahlvik and Paakkanen 2017) and teaching compassion and other character strengths in a positive psychology framework (our current study). In the other two interventions, the participants came from corporate organizations. In our interventions, the participants were only ECE head teachers. From a practical point of view, starting with head teachers is a fruitful way to foster compassion in ECEC settings. ECEC refers to the "systematic and goal-oriented entity that consists of upbringing, education and care of children and places special emphasis on pedagogy" (The Ministry of Education and Culture 2019). The head teachers are responsible for both the administration and pedagogical decision-making in their kindergarten. Most of the head teachers also work actively with children. Furthermore, the new Finnish national core curriculum for ECEC (Finnish National agency for Education 2016) underlines the role of leadership in promoting the social and emotional development of children and in creating a supportive work climate. Thus, we found it justified to train the head teachers to adopt a more compassionate working attitude as both pedagogical and organizational directors.
Educational authorities responsible for organizing and supervising early childhood education and care in a city of circa 300,000 habitants in southern Finland were contacted by the CoPassion project team (copassion.fi). They were introduced to the idea of promoting a more compassionate organizational culture in the kindergartens. This was accomplished by presenting encouraging results of studies by Dutton and her colleagues (e.g., Dutton et al. 2014). The authorities were motivated to participate in the study. They approved the study design and the protocol, that were based on the rules of ethical research outlined by the Finnish National Board on Research Integrity (2019). The educational authorities recruited the head teachers via an e-mail list, and organized the schedule and the venue for the compassion training. Participation in the intervention was mandatory and considered as part of the head teachers' yearly in-service training.

Participants
All ninety-five ECE head teachers from the city in southern Finland (94 females, mean years working as a head teacher 4.8 years) participated in our study. They were all formally competent kindergarten teachers with at least a bachelor's degree. In Finnish municipal kindergartens, the head teachers are supposed to take part in practical teaching in addition to their administrative duties. However, the weekly time that the head teachers spend with children varies a lot depending on the kindergarten size and the resources.

Study Design
A classic experimental design was applied with pre-post self-report measurements for the intervention group and a matching control group (30 head teachers from a neighboring municipality, 28 females, mean years working as a head teacher 4.2 years). Matching was based on similarities between the cities, the municipal kindergartens, and head teacher education (in Finland all ECE teachers hold at least a bachelor's degree). As far as we know, the control group was not familiar with what is meant by positive psychology nor how to use compassion and other character strengths in ECE settings.

Intervention Setting
First, 32 head teachers participated in the intervention program in Spring 2015. Then, a second set of 30 head teachers were trained in Fall 2015. Finally, the third group of 33 head teachers were enrolled in the program in Spring 2016. The contents of the program were the same each time. Based on their current workload, the participants could choose which intervention they attended. The study was completed in December 2016.

Intervention Contents
The interventions consisted of 63-hour meetings and two types of home assignments spread over 2 months. All meetings were held during the teachers' workdays in a lecture room provided by the municipality. The interventions were run by the authors who are experienced teacher trainers. The third author held only the starting lecture on the theory basis of positive psychology. The other five lectures were led by the first author. (The second author is the leader of the CoPassion project who introduced the idea to the municipality.) Participation in the intervention was mandatory. However, implementing the intervention contents in individual ECEC settings and working with the home assignments was voluntary.
Attendance at the intervention session was regarded as part of the teachers' annual work hours. Of all the participants, 86% were present at least five times. The absences were random. No specific lesson lacked participants more than the others. The absences were due to sick leave or urgent matters in the kindergarten.
The pedagogical (or andragogical) approach used in our intervention followed the framework described by Dunst (2015). The participants were introduced to new angles to guide and supervise their kindergarten practices. They were invited to combine the fresh ideas into their already existing proficiency and to form their own way of implementing new strategies in their workplaces. After the implementation, feedback was provided by the coaching specialist and the colleagues. Thus, new theory, pre-existing pedagogical philosophy, and on-going practice in the kindergarten created the medium that was evaluated and reflected during the lessons. Then, a new round of implementation was started (see also Gibbs' ideas on reflective cycles 1988).
The outline of the intervention curriculum was as follows: (1) Introduction to positive psychology, (2) What are the character strengths, (3) The power of compassion, (4) How to use compassion and other character strengths in the workplace, (5) Supportive leadership and (6) Meaning in organizations (Table 1). The meetings started with an introductory lecture about the daily topic followed by interactive assignments and practical examples in ECEC settings. The interactive assignments were prompted by challenging work life examples that the participants sought to solve together. Or, the participants stepped into the shoes of a colleague participant and helped her plan, for example, an optimal route to build a kinder communication culture. The contents of every lesson were repeated during later lessons in a spiral manner to ensure a deepening understanding of the themes. Both printed and online material were used to support the training.
After every meeting, home assignments were given based on the theme at hand. They consisted of writing exercises on one's self, in response to questions like: What do I value in myself? What are my character strengths? How could I improve my leadership skills in order to create a more compassionate work climate? The other set of the home assignments comprised practical tasks that were conducted in the kindergarten, such as giving personal feedback to staff members, paying more attention to fair division of labor, and encouraging new initiatives. The head teachers applied their fresh knowledge in their home kindergartens and furthered their knowledge by completing exercises like writing gratitude letters and diaries about their usage of strengths.
The time spent on home assignments varied from easy 15-min reflective sessions to structured observation tasks lasting several days. As is natural, great variability in the time and effort that the participants spent on their home assignment was noticed. However, the diaries revealed that all participants had at least to some extent immersed themselves in the topic and pondered on the themes that were introduced to them. Furthermore, the detailed experiences of the home assignments were shared with other intervention participants during the next meeting. These reflective discussions revealed not only enthusiasm, but also contained critical comments related to challenges such as how to find the time to help individually every staff member or how to change a deeply-rooted existing organizational culture.
After all three interventions, feedback on the curriculum and practicalities were collected. They were used as part of the rich intervention materials that were published in a book series in 2016-2019 (available in several languages, Uusitalo and Vuorinen 2020).

Analytic Strategy
In order to fully understand the outcomes of the intervention, a mixed-method approach was applied. We used both wellvalidated and new quantitative measures to explore changes in head teachers' attitudes and behaviors. Statistical analyses were carried out using SPSS 24. In addition to quantitative measures, qualitative data were collected from the third intervention group (n = 33). After the first two rounds of intervention, it became clear that finer-grained information on intervention outcomes was needed. For that purpose, the third intervention group was asked to describe their otherdirected acts as well as replying to the quantitative questionnaire. They answered open-ended questions tapping the themes of helping other people at work and how to do small acts of kindness at work. Content analysis was applied to the qualitative data.

Research Procedure and Data Collection
During the first meeting of each of the three similar interventions, the participants' informed consent was obtained and the pre-test of five quantitative self-report scales was administered. Measuring took place in a lecture hall using electronic forms. The teachers were given a link to the form and, using their own laptops, they individually responded to the research scales and the background variables of gender, age, and working experience as head teachers. Completing the pre-test (duration approximately 30 min) was overseen by the first or the third author of the study. During the last meeting, post-test was applied using the same protocol. In addition, the third intervention group answered the openended questions which took them about another 20 min.
From the original pool of teachers (n = 95), 17 did not complete both tests, but all 95 teachers that attended the intervention had completed the pre-test questionnaire. For unknown reasons, five of the pre-test questionnaires were never obtained. During the last meeting, eight participants were not present and four questionnaires were not obtained. • Exploring pathways to a more meaningful organizational culture • Group activity: reflecting what is meaningful and the core purpose of the work in a kindergarten • Homework: Acts of recognition and kindness in the kindergarten to make each worker understand their contribution in the mission and success of the organization There were no missing values in the questionnaire due to the design of the form that did not allow one to proceed without addressing all the items.

Scales
(1) Sense of compassion was measured with the Santa Clara Brief Compassion Scale (Hwang et al. 2008), which is comprised of five items (e.g., "One of the activities that provides me with the most meaning for my life is helping others in the world when they need help") rated on a seven-point Likert scale (1 = don't agree at all, 7 = completely agree). The scale was developed as a one-factor measure from the 21-item Compassionate Love Scale (Hwang et al. 2008). In the original study, the scale showed good internal consistency with a Cronbach's alpha of 0.90. In the present study, Cronbach's alpha was 0.81 in the pre-test and 0.85 in the post-test. In the study by Hwang et al. (2008, p. 426), the construct validity of the 5-item scale was demonstrated with positive correlations with the original 21-item Compassionate Love Scale and the scales of vocational identity, faith, and empathy. (2) Character strength identification was measured with a 5-item scale (Likert 1-7). The scale was modified from the original 8-item Strengths Knowledge Scale developed by Govindji and Linley (2007). The Cronbach's alpha of the 8-item scale in the Govindji and Linley study was good, 0.89. For the purposes of the current study, the items were adjusted to tap into the strength awareness in an organization (e.g., "I can see the importance of character strengths in work"). Also, some of the items were almost identical in the Finnish language and they were deleted for that reason. The internal consistency of the scale was satisfactory in the pre-test, alpha = 0.76, and good in the post-test, alpha = 0.84. (3) Strength Usage Scale (Govindji and Linley 2007) was used to measure the exploitation of strengths in the workplace (e.g., "I can use my strengths every day at work"). In the original study by Govindji and Linley (2012), the internal consistency of the 14-item scale (Likert 1-7) proved to be excellent with a Cronbach's alpha of 0.95. In the current study, the alphas were 0.92 and 0.96 in the pre-and post-test, respectively. (4) Supportive organization climate, a 10-item scale (Likert 1-7), was developed for the CoPassion project (see Paakkanen et al. 2019) in order to measure the managers' emotional skills related to the process of compassion (awareness, understanding, and acting). The scale tapped the head teachers' ability to relate to other workers in the kindergarten (e.g., "I feel I treat my colleagues fairly"). The Cronbach's alpha of the scale was excellent in both the pre-and the post-test, 0.92 and 0.95, respectively. (5) Work and Meaning Inventory (Steger et al. 2012) is a well-validated 10-item scale (Likert 1-5) probing meaning making at work and how the work is benefitting greater good (e.g., "My work helps me make sense of the world around me"). The three-factor scale proved good construct validity and excellent reliability with a Cronbach's alpha of 0.93 in the original study by Steger et al. (2012). In our current study, Cronbach's alphas were good in the pre-test, 0.85, and excellent in the post-test, 0.90.

Quantitative Results
Table 2 displays the descriptive statistics for the study variables in pre-and post-measurements. In Meaning of work, the independent samples t test showed a statistical difference between the groups in the pre-test, t(98) = 3.22, p < 0.01. In the other four scales, no differences were detected in the pre-test (ps > 0.05). The means were high, and the standard deviations were moderate in both the intervention and the control group in the pre-test.
In order to study the effects of the intervention, a repeated measures ANOVA was performed (Table 3). Time and group (intervention/control) were entered as independent variables for each of the five scales. Greenhouse-Geisser correction was applied when the assumption of sphericity was violated.
Time had a major effect in four of the scales but only the intervention group made statistically significant progress between the pre-and the post-measurement. In  (Table 3).

Qualitative Methods
Participants in the third intervention (n = 33) replied to open-ended questions after finishing the intervention. Specifically, they were asked to write about their otherdirected acts and strivings with the following two prompts: How does your daily work help other people? What small acts of kindness could you do more often? These prompts were based on the practical home assignments and discussions that arose from them during all three interventions. Moreover, the abundant ideas about acts of kindness uttered by the participants were considered to be worth collecting for direct practical use (they have been published in our book series, Uusitalo and Vuorinen 2020).
Two researchers independently read the accounts. Memos were written during the initial reading and the coding process to ensure reliability of the analyzing process. The 99 pieces of text (two to four answers from each respondent, average length two sentences), were studied using inductive content analysis (Elo and Kyngäs 2008). The analysis proceeded in three phases. First, the 99 text pieces were reduced to codes. Initially, in vivo coding was applied (see, Manning 2017). To let the respondents' actual voices to be heard, words extracted directly from the text pieces were used as a starting point (e.g., the word presence existed in a multitude of accounts). Then, axial coding followed. The codes were based on the data, and every time a new topic emerged, a new code was created. The coding phase included multiple readings of the data. Statements that included many kinds of codes were divided into several categories. After coding the data, sub-categories and generic categories were formed to give three main categories, namely Caring professionalism, Warm presence and Positive feedback. The main categories were named according to the content of the subcategories and generic categories (Elo and Kyngäs 2008). An example of the categorization process is given in Fig. 1.
In the accounts, oft-repeated phrases were every encounter matters (main category Warm presence), a genuine presence (main category Warm presence), trying to be fair (main category Caring professionalism). Creating good working conditions was directly mentioned in 13 accounts.
Through change in work culture well-being will be reached. (R23, main category Caring professionalism) Striving toward a considerate and supportive spirit was voiced in 12 accounts.
I hope the people you meet would feel that you respect them. (R21, main category Positive feedback) Eleven of the respondents wanted to be able to give more encouraging feedback and to notice even the smallest positive acts.
I could use more words to thank others for their kind deeds. (R20, main category Positive feedback) I could give more immediate feedback. (R3, main category Positive feedback) Although a compassionate mindset was evident in many accounts, compassion was precisely mentioned in only three accounts, such as. To summarize the qualitative findings, it can be concluded that the ECE head teachers saw their role as guarantors of a safe workplace trying to find place for more individualized, positive feedback. The head teachers were very aware of their responsibility to maintain and promote a spirit of mutual understanding in their kindergartens. Taking care of ongoing issues, securing circumstances, and backing up daily functions seemed to be a source of stress for many respondents. Only the concept of managing human resources and administration was indicated as being not what the professional early childhood educators had aimed at in their career choices. Their roles as pedagogical leaders seemed to fade in the numerous organizing tasks in the daily hassle. The children and the salience of showing them a good example of a kind and compassionate adult were not mentioned in many answers. Perhaps the Fig. 1 The categorization path to form the main category of Caring professionalism questions and the context led the responses in that direction, or perhaps the head teachers truly considered themselves to be more staff leaders than practical pedagogues. However, all head teachers are educated early childhood experts and their participation in interacting directly with the children is highly recommended.

Discussion
The purpose of this study was to investigate the outcomes of a compassion training for early childhood education (ECE) head teachers in Finnish kindergartens. Our research is part of a large compassion-promoting project called CoPassion (CoPassion.fi), which has produced several interventions for a variety of organizations (Hakanen and Pessi 2018;Paakkanen et al. 2019). These and several other studies (Gilbert and Procter 2006;Kirby 2017;Lilius et al. 2011;Niemiec 2017) have shown that compassion is a character skill that can be acquired through training in a variety of settings. In the new Finnish national core curriculum for ECEC (Finnish National agency for Education 2016) the role of leadership is underlined in promoting social and emotional development of children and in creating a supportive work climate. So far, not only is training for ECE leaders in these skills scarce, but few attempts have been made to study them. The aim of our current study was to fill this gap. In our three similar interventions, 95 head teachers participated in 63-hour training sessions with pre-and post-measurements. In addition, qualitative data were collected from 33 participants.
We were interested in finding out how compassionate behavior, strength identification, and strength usage could be nourished to create a more supportive work culture. In our interventions, compassion and other character strengths, supportive leadership and meaning in a workplace were explicitly taught. The lessons were followed by home assignments and practical tasks that the head teachers carried out in their kindergartens (such as writing about valuable things in oneself and giving more positive feedback to their staff). ECE head teachers from a neighboring municipality served as a control group.

Compassionate Behavior
Promoting compassionate behavior in ECE head teachers was the main aim of our study. Compassion skills were taught alongside other "strengths of heart" as a human capability aimed at alleviating pain in others. We believe that we touched only the surface of the topic in our two-month training period. Nevertheless, merely raising awareness of the theme seemed to bring results. As could be expected, ECE head teachers' attitudes were other-oriented and already helpful before the intervention. Both the intervention and the control group scored high in the Sense of compassion scale comprising questions on their other-oriented, empathetic feelings. However, no ceiling effect could be detected. The intervention group scored significantly higher in the post-test than in the pre-test. No such progress was visible in the control group. It was our view that the core ingredients of our intervention (making the concept of compassion more tangible and paying attention to small acts of kindness), sharpened teachers' attitudes to alleviating suffering in others. Also, joy and success seemed to be noticed more often.
As written by Dutton et al. (2014), compassion is a manyfold process which starts from noticing the pain in others. The municipality from which our participant teachers came consists of areas with very different socio-economic structures. In some areas, challenges with children and their families are substantial and the workload, and presumably the work-related stress, of staff members is heavy there (see also Rajala et al. 2019). However, personal life events may cause pain and suffering in all working conditions, and the number of burdening issues in the kindergarten staff, the children or their custodians was not known. Nevertheless, we were delighted to witness the start of participants paying more attention to others' emotions and being prepared to support each other. The results of our mixed method study show that even with a short intervention, it is possible to spark the idea of a more compassionate organization culture. The qualitative data confirmed that the ECE head teachers wished to have a closer eye on their staff's concerns and to be vigilant enough to help them.

A New Definition for Compassion
Based on our current results (see also Hakanen and Pessi 2018;Paakkanen et al. 2019), we suggest that the definition of compassion should be extended to encompass co-joy and co-enthusiasm alongside co-suffering (see https ://copas sion. fi/en/compa ssion /). Since the Latin passio also means strong emotion and desire, this interpretation is reasonable. In our qualitative accounts, the respondents frequently mentioned the need to express their gratitude, respect, and admiration to the staff members. In the response categories of Warm presence and Positive feedback, willingness to encourage, to cocelebrate and make acts of kindness more often was evident. It was seen to be as important as noticing the pain in others and sometimes even harder since the positive moments seemed to vanish in the daily routines or they were taken too much for granted.
Thus, our suggestion for a new definition of compassion includes noticing, feeling, and acting to alleviate the pain and to reinforce the joy in another person. Our definition of compassion is an antidote and reducer of the mental carbon footprint, the concept of ignorant, negative and oppressive behavior that loosens the ties between people, increases competition and ultimately decreases subjective well-being in everyone ; see also, posit ive.fi).

Strengths Identification and Usage
Our interventions were based on positive psychology and the definition of character strengths as the main building block of well-being and flourishing (Seligman 2012;Vuorinen et al., in press;Wagner et al. 2019). Strength identification means learning about the concepts of different strengths and being able to name them. Strength usage taps the capability to intentionally exploit skills in your character (Govindji and Linley 2007). In our interventions, the idea of what is meant by character strengths and what kind of character strengths exist were the starting point. All 24 character strengths categorized by Peterson and Seligman (2004, see also viacharacter.org/research) were introduced with some modifications when we conducted our Finnish studies (Vuorinen et al. 2019, in press). The emphasis was on strengths that represent the virtue of humanity, namely kindness, love, and social intelligence. Since the current study was started in 2015, we have separated compassion from kindness and given it the status of a character strength per se, not just a sub constituent or relative of kindness. In our later studies and teacher training materials (Kotilainen et al. 2020;, compassion exists as one of the 26 strengths encompassing similar strengths to the VIA character strengths and two additional strengths, namely compassion and sisu (Finnish grit, see Lahti 2019).
Strength identification and usage are intertwined and affect each other. The more you become aware of your own (and others') strengths, the more likely you are to use them. Our present interventions brought promising results in both strength identification and usage. There was a significant interaction effect (with medium to large effect sizes) of time and group, in favor of the intervention group in both the identification and the usage of strengths scales suggesting that the strengths training had a positive effect. As in previous studies, even a short intervention can result in significant gains in general character strength knowledge (Niemiec 2017;Wagner et al. 2019;Vuorinen et al. 2019). Having a profound understanding of specific strengths and how to harness them to one's benefit seems to be a longer lasting endeavor (e.g., Seligman 2012). Only one respondent mentioned the word "strength" in the qualitative accounts. However, intentions to pursue a more strength-centered organization culture were evident.

Creating a Meaningful and Supportive Work Culture
Paying attention to others' suffering and trying to understand and act in order to ease the pain can pave the way for a community in which competition is replaced by compassion (Lilius et al. 2008). At the heart of the current study was the idea of nourishing compassionate behavior in ECE head teachers, to create a supportive work culture. Both the quantitative and the qualitative data gave encouraging results. The intervention group made progress over the control group on both quantitative scales, Meaning of work and Supportive organization climate, aimed at measuring the emotional climate and social relationships in the workplace. In Supportive organization climate, there was a statistically significant interaction effect showing the progress of the intervention group.
The qualitative accounts revealed increased awareness and willingness to give more frequent positive feedback and to continue listening to people's ideas and challenges. The words respect and appreciation were present in many accounts. Finding meaning in work through binding people together, showing them trust and giving praise were uttered by several respondents. The three central themes of Caring professionalism, Warm presence, and Positive feedback categorized from the qualitative data describe the idea of a supportive work culture and the head teachers' striving toward it.

Limitations
Each research design has its own limitations and this study is no exception. First, the head teachers were recruited by the municipality and participation was not voluntary but considered to be part of their work duties. Not all participants were enthusiastic about starting the training. This may have affected some of the results. On the other hand, our sample represented the head teachers of the area well. The educational authorities had no access to the primary research data, which reduced the chances of giving socially desired answers. The overall feedback from the compassion training (that was not used as research data) was frank and positive. Comments on the intervention aims, contents, methods, exercises, and practicalities encouraged us to produce materials (Uusitalo and Vuorinen 2020, see also posit ive.fi) and develop ECE teacher training further.
Second, the number of participants in the control group was significantly lower than the number of participants in the intervention group. For the analysis of variance, the size of the control group was enough, though. Third, the Supportive organization climate scale had been used only once before our current study (Paakkanen et al. 2019), and the psychometric properties of the scale have not been thoroughly analyzed. Nevertheless, the internal consistency of the scale proved to be excellent. Fourth, the contents of the interventions were not identical but changed slightly due to the development of the exercises and the desires and needs of the participants. This was seen as a benefit to the outcomes and served the teachers better. Fifth, the kindergarten children (or their families) were not included in the study. Their voices would have complemented the data and given real life experiences of how compassion among adults shows in their eyes. This angle should be included in the future studies.

Implications
In order to make visible the full impact of the intervention, behaviors of all actors in the ECEC settings should be monitored. This includes all the staff and their interaction with the children and the custodians. However, we consider training those in charge of the pedagogy and the administration of an ECEC organization to be a necessary starting point. In the future, what happens in the ECEC settings ought to be monitored with care. Self-reports are only the first step; observations and ethnographic research are needed as well. Ultimately, one of the core aims of early childhood education is to nurture children's prosocial skills (e.g., Taggart 2016). As written by Spinrad and Eisenberg (2017, p. 60), "interventions to develop children's compassionate behavior is a long-term goal and understanding the mechanisms involved in effective interventions is a key topic for future studies".
Overall, we found our results to be promising and giving a firm starting point for relevant compassion and character strength training in ECEC settings. As practical implications of the current study, we can conclude that there was high demand for training for in-service ECE teachers, as well as for student teachers to adopt a more other-oriented, actively compassionate role in their work. Training the head teachers is a necessary start to a new pedagogy in ECEC settings. The head teachers are responsible for the curriculum and daily functions in their kindergartens. It is up to them to spread the knowledge of a kinder, more empathic social culture that in the end greatly benefits the children's ethical development and feelings of safety (Taggart 2016). If the head teachers are not committed, a change in the ECEC culture will be hard to implement. However, all the staff need to be trained to ensure a common appreciation of mutual compassion and reduced competition in their own roles as educators.
As the most important element of the compassion training, we consider understanding that learning to share both pain and joy can empower the whole organization and improve its functions in many levels. Being left alone and feeling unsupported by colleagues or managers is one of the greatest sources of stress in many workplaces, not only in ECEC settings and schools (e.g., Hakanen et al. 2006).
Since the start of our journey in compassion interventions in 2015-2016, thousands of teachers all over the world have been trained to identify and use compassion, kindness, and other character strengths in their workplaces. Our printed and digital materials have been translated into five languages, and several new interventions have been conducted , see also posit ive.fi).
It takes time and practice to adopt a whole new approach in social interactions. However, an attitudinal change can lead to small acts of compassion that are visible in everyday life. As written by one of our participants: One encounter can make all the difference.