Emotional competencies in emerging adolescence: relations between teacher ratings and student self-reports

Abstract This cross-sectional study examines gender-informed teachers’ and early adolescents’ reports of emotional competencies within the school setting. The sample for the current study consisted of 290 emerging adolescents (114 boys, 170 girls) recruited from 24 schools. Social-emotional competencies were assessed by students’ self-report questionnaires and teacher and parent reports. Results showed significant positive associations between teacher-rated emotional competencies and emerging adolescents’ self-reports with three of the four Interpersonal Reactivity Index subscales (Perspective Taking, r = .231, p = .000; Fantasy, r = .303, p = .000; and Empathetic Concern, r = .196, p = .007). Significant gender differences were also found as both adults and adolescents’ rated girls as showing greater emotional competencies than boys. Results are discussed in terms of future research directions and implications for educational strategies and practices to help build prosocial and emotional abilities and behaviours.

Emotional competencies include knowledge and skills necessary to understand and manage emotions, set and achieve goals, feel and show empathy and compassion for others, as well as establish and keep positive relationships (Weissberg, Durlak, Domitrovich, & Gullotta, 2015). Emotional competencies among children and adolescents have also been found to be associated with social, behavioural and academic outcomes (Domitrovich, Durlak, Staley, & Weissberg, 2017). Although an association with academic outcomes exists, there are limited findings between teachers' ratings and students' self-reports of emotional competencies such as empathy and achievement. To the best of our knowledge, there remains a lack of studies that explicitly examine relations between teacher-and self-reported emotional competencies of adolescents within the classroom. The present study aimed to explore this particular relation between teacher and student self-reports, with a focus on gender differences.
Teachers have the ability to promote positive relationships amongst peers in the classroom, which is a large part of student learning and development (Audley-Piotrowski, Singer, & Patterson, 2015). Teachers can assist in this process through reinforcing classroom norms as well as positive behaviors, thereby increasing positive peer relationships (Audley-Piotrowski et al., 2015). Therefore, teacher's understanding of children's emotional competencies is a vital component of a well-functioning classroom, as it promotes supportive and compassionate teacher-student relationships and helps to create an effective learning environment for students. Knowledge about students' emotional health and abilities is especially important during the transition from later childhood to early adolescence, or the period of ARTICLE HISTORY received 5 february 2018 accepted 17 March 2018 KEYWORDS emerging adolescence; theory of mind; emotional competence; teacher-reports emerging adolescence (ages 9-13 years) because emerging adolescence has been found to be a time of change in all areas of development including social interactions and physical, cognitive and affective growth (Ellis & Zarbatany, 2017;Greenspan & Deardorff, 2014). Therefore, the knowledge educators possess regarding their students' emotional competencies may help to them develop teaching and classroom resources and routines that can benefit students' emotional well-being.

Emotional competencies in adolescence
Throughout literature the term 'emotional competencies' has been used to represent various constructs that range from general abilities, such as emotional understanding (EU), to specific emotion-related skills, such as emotional intelligence (Keefer, 2015). For example, Emotional Intelligence (EI) refers to using emotions to expedite thinking, understanding and managing of one's emotions, and can otherwise be seen as the perception of emotions (Poulou, 2016a). 'Perception of emotions' needs clarification here-does it mean emerging adolescent perception of other people's emotions? Similarly, Theory of Mind (ToM) is the ability to understand mental states such as emotions, desires and beliefs, and then using this understanding to predict the behaviors of others as well as one's self (Valle, Massaro, Castelli, & Marchetti, 2015). Although past research suggests that ToM and EU are interrelated (Preckel, Kanske, & Singer, 2017), studies have also found that these social-cognitive abilities may also be distinctive of one another (Devine, White, Ensor, & Hughes, 2016;Kuhnert, Begeer, Fink, & Rosnay, 2016). Such skills become particularly present during middle childhood, when children develop their ability to interpret social rules and react to their own and other people's emotional cues according to the social and cultural context (Kuhnert et al., 2016).
Empathy represents an important emotional skill that contributes to the development of adolescent's socio-emotional abilities and experiences. This includes Tom, prosocial behaviors, and externalizing (social or interpersonal) and internalizing (personal or intrapersonal) challenges (Sánchez-Pérez, Fuentes, Joilliffe, & González-Salinas, 2014). For example, Lam, Solmeyer, and McHale (2012) examined empathy's course of development from seven to 14 years of age, using a multilevel modeling strategy that charted the growth of empathy. Their results demonstrated that brain areas involved in social cognitive processing reflected major changes during adolescence. More specifically, they found that peer interactions allowed for opportunities practice interpersonal negotiation skills and thus helped with internalizing and externalizing challenges (Lam et al., 2012). This suggests that empathy plays a role in self-perceptions as well as social behavior and is important for teachers of teenagers to be aware of.
In addition, within the social domain, as children approach adolescence, gender differences become prevalent and intensify into adolescence (Greenspan & Deardorff, 2014). For instance, past studies have shown that compared to girls, boys primarily socialize in larger same-sex peer groups that focus on competiveness (Rose & Asher, 2017). In contrast, girls have been found to typically intermingle with smaller-number groups of same-sex peers that are more intimate and collaborative (Kuhnert et al., 2016).
More specifically, Kuhnert et al. (2016) found that for both boys and girls, earlier ToM was found to be important for later prosocial behavior. In particular, compared to boys EU was predictive of prosocial behaviour for girls. Patterns of change in empathy also varied between girls and boys during their transition into adolescence, where girls showed an increase in empathy during the transition and reached a plateau around thirteen years of age (Lam et al., 2012). Boys also had a temporary increase in empathy during their transition to adolescence followed by a decline in empathy by the age of 13 years (Lam et al., 2012). Such gender-related differences in the development of empathy in part supports the gender-intensification hypothesis that claims that the pressure of gender-role stereotypes and expectations increase during adolescence (Lam et al., 2012;Wright, Rieldel, Sechrest, Lane, & Smith, 2018).

Measurement of emotional competencies
Children's self-report competence ratings have been widely used in psycho-educational practice for over fifty years and remain one of the most commonly used assessment tool of emotion-related constructs (Keefer, 2015). A variety of methods, such as direct and structured observations of behaviors, have been used to measure empathy. Although other methods exist, self-reports are the largest type of measure used when studying empathy, as it requires the least amount of investment while still delivering comprehensive information (Sánchez-Pérez et al., 2014). For example, one of the most widely-used and validated assessment of children's emotional competencies is The Interpersonal Reactivity Index (IRI) which is a self-report scale used to measure individual differences in empathy.
As young people continue to develop cognitively, their emotional competencies become more complex (Keefer, 2015). Keefer (2015) documented the amount of shared variance from self-reports on multiple emotional intelligence abilities, demonstrating a large decrease from 26 to 5% during ages 10-11 years and age 16. This suggests that in addition to adolescents' self-reports, reports from significant others such as teachers and parents are needed to reflect an accurate picture of adolescents' social and emotional skills. Therefore, the use of teacher reports would be a valuable addition to this type of research.
Teacher ratings of student behaviours have been reported to have greater validity in relation to parents' ratings, specifically in areas such as school outcomes (Pas & Bradshaw, 2014). A large amount of research exists to explain the importance of teacher-student relationships; however, teachers' own views of social and emotional skills and the effect they may have on the classroom are largely overlooked (Poulou, 2016a). There is limited knowledge in regard to how teacher-student interactions relate to the teacher's emotional understanding (Poulou, 2016a). Poulou (2016b) says that the experiences teachers provide within their classrooms assist their students to promote their own emotional competencies. Teacher ratings of children's social competence in kindergarten when compared to children's self-reports of connectedness to others in middle childhood, showed to be a strong predictor of emotional competencies (Guhn, Gadermann, Almas, Schonert-Reichl, & Hertzman, 2016). Over time teacher ratings were stable, even when data was being collected over various school years and were on assorted cohorts of students (Pas & Bradshaw, 2014) suggesting that teacher ratings of students' social and emotional competencies may be valuable and add to research on young people's social cognition and behavior.
Past research also shows mixed findings regarding gender-related differences among children's emotional competencies (Wright et al., 2018). Differences have been found as young as 10 years of age (Keefer, 2015). For example, when being rated by others, studies have found that compared to boys, girls tend to engage in more prosocial acts towards their peers (Ellis & Zarbatany, 2017;Kuhnert et al., 2016). Sánchez-Pérez et al. (2014) found that self-reports of boys and girls showed differences in affective empathy, with girls' self-report scores tending to being higher than boys suggesting that differences may exist in boys' and girls' perceptions of their emotional states.
In contrast, Sánchez-Pérez et al. (2014) found no gender differences within parent reports of their children's empathic skills. In addition, after controlling for differences due to biological sex, Keefer (2015) found gender relations in empathy did not differ, and remained substantial. Thus, given such contradicting results, more research on the relations between gender and empathy is needed (Wright et al., 2018), especially during the transition to adolescence when societal views of gendered stereotypes may play a role in adolescents' self-perception (Keefer, 2015) (Guhn et al., 2016;Keefer, 2015).

The present study
Thus, building on the aforementioned literature above, the present study will explore the relations between adult ratings and student self-report of emotional competencies among young adolescents. Given that to the best of our knowledge, there remain a lack of studies that explore gender-differences among parent, teacher and student reports of their emotional competencies during emerging adolescence, the purpose of this study was to the examine gender-informed teachers' and children's reports of emotional competencies within the school setting.

Participants
As part of a larger study on ToM in emerging adolescents, the current study represents data from Year 1 (2015-2016). The sample for the current study included 290 emerging adolescents (114 boys, 170 girls; mean age = 12.7 years, ranging from 9.9-14.3 years), as well as 184 teachers and 61 parents. The majority of participants were recruited from 24 schools from mainly English-speaking, middle socioeconomic status populations, while the remaining participants were recruited from university research lab participants. The majority of the participants (245/290) had either a teacher (if recruited from a school), or a parent (if recruited in the lab) complete a questionnaire about the participant (see Table 1).

Teacher and parent reports
Teachers of participants recruited in schools and parents of additional participants recruited in the lab completed the Children's Emotional Competence Scale (CECS), adapted from Denham (1998). This 10 item measure assesses children's emotional behaviours and competencies, including the ability to read and understand emotions, react to and regulate emotions. Teachers and parents were asked to rate the emerging adolescents on a five-point Likert scale (1 = this is never true of this child, 5 = this is almost always true of this child). Examples of items from this measure include 'this child demonstrates empathetic involvement in others' emotions' and 'this child understands one's own emotional states' . Participants were only given a total score for the CECS if their teacher or parent completed at least 5 of the 10 items. A mean score of the 10 items was calculated for the total score.

Student self-report
Adolescent participants completed the Interpersonal Reactivity Index (IRI). This measure is made up of 28 items which can be divided into four subscales that are made up of seven items each, all measuring a different aspect of empathy (see Table 2). All items were scored on a five-point Likert scale (1 = does not describe me well, 5 = describes me very well). The IRI showed good internal reliability (.71-.77) and good test-retest reliability (.62-.71) for all four subscales (Davis, 1983). 'I really get involved with the feelings of the characters in a novel. ' empathetic concern ability to feel empathy for others 'I often have tender, concerned feelings for people less fortunate than me. ' Personal distress negative feelings one might feel in reaction to others' emotions 'In emergency situations, I feel apprehensive and ill-at-ease. ' Subscale scores were calculated using the mean for the seven items within that subscale. For example, Perspective Taking score was calculated using the mean of the Likert Scale responses to all seven-statement items which describe various aspects of perspective taking. To receive a score on a particular subscale, participants needed to obtain a score of at least four out of the seven items, meaning four items must have had an appropriate Likert Scale response given in order for that subscale to receive a calculated mean score.

Procedures
As part of a larger five-year longitudinal study on emerging adolescents' ToM, relationships, and sense of identity, upon receiving university and school board ethical clearance, the data used in the current study was taken from the first year of this longitudinal study. Once written, informed consent was received from, teachers or parents, a paper and pencil version of the CECS was administered to participants on the first school or lab visit. These measures were then collected on the final school or lab visit by a research assistant, or in some cases were mailed back to the research lab. Once research assistants received students' verbal and written assent, they administered paper and pencil versions of the IRI to groups of participants and instructed participants to complete the measure on their own. Participants completed this task on the first data collection visit along with other measures that were part of the longitudinal study.

Preliminary analyses
Means and standard variations for all variables were normally distributed (see Table 3). There were also six outliers in the data-set (i.e. one in the teacher reports and five in the self-reports). The results presented below are a product of analyses conducted without these six outliers.

Correlations between teacher and student reports
To determine the relations between teacher and student self-reports of emotional competence, twotailed correlations were conducted between parent and teacher rated CECS scores and the four subscales of the self-reported IRI. Teacher-rated CECS scores were significantly positively correlated with three of the four IRI subscales (Perspective Taking, r = .231, p = .000; Fantasy, r = .303, p = .000; and Empathetic Concern, r = .196, p = .007). Furthermore, parent-rated CECS scores were significantly positively correlated with the Fantasy subscale of the IRI, r = .328, p = .009. This suggests that emerging adolescents who scored higher on teacher-rated emotional competence were more likely see themselves as having the abilities to understand someone else's view point (Perspective Taking) and feel empathy for others (Empathetic Concern). Moreover, this also suggests that emerging adolescents who scored higher on teacher-rated emotional competence were more likely to feel that they were good at putting themselves in the shoes of fictional characters (Fantasy). Teacher-rated empathetic involvement was significantly positively correlated with all four IRI subscales (Perspective Taking, r = .231, p < .001; Fantasy, r = .206, p = .001; Empathetic Concern, r = .269, p < .001; and Personal Distress, r = .150, p = .019). This suggests that emerging adolescents rated by their teacher or parent as having higher empathetic concern were more likely to rate themselves as having empathetic abilities in all four areas measured by the IRI.

Gender differences
Independent samples t-tests were conducted to determine whether there were significant gender differences in how adults rated emerging adolescents' emotional competence, with CECS scores as the dependent variable and participant gender as the independent variable. Significant gender differences were found for both teacher and parent-rated emotional competence, t(180) -6.052, p = .000 and t(59) = 2.284, p = .026, respectively. Means and standard deviations for CECS scores can be seen in Table 4. Both teachers and parents rated girls as having greater emotional competence than boys.
Independent samples t-tests were also conducted to determine whether there were significant gender differences in how emerging adolescents' rated their own emotional competence. Significant gender differences were found for all four subscales of the self-reported IRI (Perspective Taking, t(271) = 2.815, p = .005; Fantasy, t(287) = 3.367, p = .001; Empathetic Concern, t(286) = 4.230, p = .000; and Personal Distress, t(280) = 4.374, p = .000). Means and standard deviations for IRI scores can also be seen in Table 4. Girls rated themselves as having greater empathetic abilities than did boys in terms of understanding someone else's point of view, putting oneself in a fictional characters' shoes, feeling empathy for others, and feeling negative emotions in the face of someone else's emotions.

Discussion
The purpose of the current study was to explore the relations between adult and self-reported emotional competence in emerging adolescents. Furthermore, gender differences in adults' and emerging adolescents' emotional competencies were also explored. Significant associations were found between teacher-rated emotional competencies and emerging adolescents' self-reports of the Perspective Taking, Fantasy and Empathetic Concern subscales of the IRI. Parent-rated emotional competencies were also significantly related to self-reported Empathetic Concern. Additionally, teacher-rated empathy was significantly related to all four areas of emerging adolescents' self-reported empathy. Subsequently, both adults and emerging adolescents' rated girls as showing higher emotional competencies than boys. Each of these findings will be discussed below within the context of past literature.
The present study's findings regarding gender differences in emotional competencies supports past research on the role gender plays in adolescents' as well as adults' social and emotional abilities (Wright et al., 2018). For example, Karniol, Habay, Ochion, and Harari (1998) examined how gender and gender-role orientation (perceptions of stereotypic femininity and masculinity) contributed to empathy in adolescence. Karniol et al. (1998) found that boys obtained a lower overall IRI as well as empathy scores when compared to girls. Individuals with high femininity had overall IRI and empathy scores higher than those who were low in femininity (Karniol et al., 1998). In addition to gender, Karniol et al. 's (1998) study found a high correlation between empathy and self-perceived femininity, however there was no negative correlation present between self-perceived masculinity and empathy. This demonstrates that both gender and gender-role orientation contributed to empathy (Karniol et al., 1998). Although the present study did not investigate adolescents' gender-role self-perceptions, the present findings suggest that gender does play a role in emotional competencies, especially in empathy. Prior research and the current findings may provide ideas for future research to explore the role biological sex and gender-role perceptions play in adolescents' self-reports and teacher as well as parent reports of their social and emotional behaviors.
In contrast to the present findings, Sánchez-Pérez et al. (2014) found that there were no gender differences in self-reported cognitive empathy, as parent reports showed no gender effects in any of the empathy scales. Parent reports on cognitive empathy also did not increase with their child's age, which the authors claimed may have been due to the diverse situations in which the parents observed their child's behaviors. However, this difference may have been because Sánchez-Pérez et al. (2014) examined parent reports over time, whereas the current study focused on parent and teacher reports from one time only.
The significant results regarding the positive correlations between teacher-rated and adolescents' self-reported empathy might have been due to the lack of familiar contexts that teachers observe. Teachers observed their students under a variety of circumstances, having more time with them in their awake hours than their parents in some situations, which may have provided them with more accurate observation of their emotional competences. Past research shows that gender-related changes in empathy patterns occur over time and often result in a widened gap through the transition to adolescence (Lam et al., 2012).
Building on past studies that show links between compassion and empathy among youth (Neff & McGhee, 2010), Bluth, Campo, Futch, and Gaylord (2017) examined levels of self-compassion across adolescence, and their association with indicators of emotional well-being. Specifically, this research pursued how self-compassion differed by age and gender, including its associations with emotional well-being (Bluth et al., 2017). They hypothesized that self-compassion would be lower around adolescents due to the co morbid internalizing symptoms that increase around this time. Bluth et al. (2017) found that there were gender and age differences in self-compassion with males holding a similar level across all ages, while the older females had the lowest levels of self-compassion. This research supports the idea that many internalizing factors change over the adolescent period, demonstrating that adolescence is a sensitive time of multiple developmental transitions. Thus, early adolescence needs to be researched further to help researchers and teachers understand the complexity of these psychosocial changes and how the immediate people adolescents' lives view them.

Limitations and future directions
Certain limitations should be kept in mind when considering the results of the current study. First, a co relational design was used and thus, causation or directionality cannot be inferred in the significant relations reported. Future research should include a cross-cultural, longitudinal design to determine the direction of the directional relations found across ages, gender, and whether cultural heritage or ethnicity is a factor is ToM and empathy. As the current data is a snapshot of data from one year of a larger five-year study. Therefore, it would be beneficial to improve on the current study once the longitudinal study is complete for a clearer picture of how teacher reports of young adolescents' social behavior may predict their self-reported social behavior in later adolescence (or vice versa) between adult-and self-reported emotional competence.
Second, the adult-reports measured a wider scope of variables related to emotional competence than did the self-reports, which focused on empathy. Future studies could benefit from giving adapted versions of the same measures to both adults and emerging adolescents to gain a more accurate comparison between adult and self-reports. Finally, although significant, many of the correlations reported in this study were relatively small, the result may have been due to the difference in adult and self-reported measures mentioned above.

Implications for education
The current study has implications for developmental and educational research in young adolescents' social cognition and well-being. Theoretically, this study advances our current knowledge on how emerging adolescents view their own emotional competence and highlights any similarities and discrepancies between how teachers and parents view emerging adolescents' emotional competence and how emerging adolescents' view their own emotional competence. The present findings also add to the current research on how young teenagers view themselves, as compared to childhood and young adults, emerging adolescence remains age group that is often overlooked research on identity and social emotional factors.
Practically, the current study offers implications for educational settings. Given that the present study suggest that there remains some discrepancy between how teachers and students view their emotional competencies, daily classroom discussions should include dialogue on emotional understanding skills such as empathy and compassion (self and other). For example, teachers could include daily discussions of emotional experiences students would like to share, and integrate mental state talk into daily teaching practice sand curriculums. Serving as role models, teachers have the unique opportunity to encourage emerging adolescents' to develop their empathetic skills and overall emotional competence. Inclusive, gender and developmentally sensitive cooperative and dialogic activities such as classroom discussions of current newspaper headlines that focus on well-being and social justice issues could promote the ability of students to regulate and monitor their emotions within the classroom on a daily basis.
Moreover, the current study also highlights the need to focus on these skills for emerging male adolescents' who, based on both teacher and self-reports, may benefit from programs that target empathy, compassion, and connections to prosocial behaviours (Neff & McGhee, 2010;Way, 2011). For example, teachers could encourage male students to talk about their feelings about their pets, friendships or how they feel about winning or losing in sports. Such life skills are crucial for early adolescents of all gender orientations and should be a main focus of high school social-emotional skill and moral reasoning programs.
Keeley Dutcher is a graduate student pursuing Master in Social Work with research interests in youth mental health and justice.
Malak Askar is a Research Specialist at Hamilton-Wentworth District School Board with a portfolio of culture, diversity and well-being research projects.