The initial search yielded 3,377 records in total. A total of 3,360 records found from the four electronic search databases (2992 from Google Scholar, 183 from PubMed, 184 from ScienceDirect, and 1 from Ebscohost), and 17 studies were found by reviewing the reference lists and citation searching. There were 3,117 records after the duplicates were removed. Following the screening of title and abstracts for studies, 117 studies were selected for further evaluation. After further review and application of selection criteria, 45 studies were selected for inclusion in this review. A PRISMA flow diagram is prepared to illustrate the study selection process (see figure 1).
The extracted data were summarised by the studies' country, the type of study design, and the social-emotional domains studied. From the extracted data, six themes emerge; which is (1) Social-emotional development; (2) Social competence; (3) Emotional development; (4) Problem behaviour; (5) Self-regulation; (6) Social-emotional learning. There was one study that could not be categorised as it coded into more than one category. Three studies were collated into one category as they measured the same social-emotional domain but differed in the definition used.
Summary of the included study characteristics
The majority of the studies were published in 2020 (n=11, 24.4%), followed by 7 (15.6%) studies in 2018 and 2019 and 6 (13.3%) studies in 2017. From 2010 to 2016, one to four (2.2 – 8.9%) studies were published annually (see Figure 2). A total of 46,625 participants involved in these social-emotional studies were children with typical development, parents or teachers, and 689 involved children with disabilities (physical or developmental). A vast majority of these studies involved only children as an informant 18 (40.0%). Followed by 17 studies (37.8%) involved either parent or teacher, 5 studies (11.1%) included all three as informants (parent, teacher, child), 3 studies (6.7%) involved teacher only, and only 2 studies (4.4%) involved the parent. Table 1 presents a summary of the characteristics of the included studies and their key findings.
Summary of social-emotional research in Asian countries
All 45 studies included for qualitative synthesis were primary studies on social-emotional development published in English and were within the Asia continent. Out of the 45 studies, 13 studies originated from China, 7 studies from Hong Kong (HK), 3 studies each from South Korea, Malaysia, Singapore, Thailand, and Turkey, and 1 study from each of India, Japan, Pakistan, and Taiwan. Two studies compared social-emotional development in two countries; (India & China, US & South Korea) (see Figure 3).
Summary of the type of study design
Among the 45 studies, 23 (51.1%) were cross-sectional, 5 (11.1%) interventional studies, 4 (8.9%) each of validation studies and longitudinal studies, 3 (6.7%) each of mixed-method studies and cohort studies, 2 (4.4%) experimental studies, and 1 (2.2%) observational study (see Figure 4).
Summary of the social-emotional domain studied
The vast majority of the studies focused on overall social-emotional development domain (n = 24, 53.3%), followed by specific social competence domain (n = 7, 15.6%), emotional development (n = 5, 11.1%), social-emotional learning (n = 3, 6.7%), problem behaviour (n = 3, 6.7%), self-regulation (n = 2, 4.4%), and 1 study focused on social-emotional learning and problem behaviour (n = 1, 2.2%) (see Figure 5).
Summary of Social-Emotional Research Key Findings
This systematic review coded the included studies according to the social-emotional development domains assessed. Overall social-emotional development is the most studied domain (n = 24, 53.3%) among the 45 studies reviewed. Overall social-emotional development in this systematic review is referred to studies that do not specify a specific construct within social-emotional development. There are nine studies conducted in China that examined the overall social-emotional development. From the nine studies conducted in China, five studies examined the association of social-emotional development and children’s academic skills [13, 26-29].
The remaining three studies were on the family structure and home environment [30-32]. One study developed a social-emotional development scale for the Chinese population to measure boarding schools' effect on left-behind children and their development .
Out of these five studies, most of the findings were consistent with other studies whereby there is a negative correlation between children’s social-emotional development and their academic skills. A study conducted by Ren et al.  found that father’s parenting is not a mediator to the outcomes of children’s social-emotional development and academic skills. This finding is inconsistent with the studies from other parts of the world whereby parenting was reported to be one of the mediators in children’s social-emotional functioning and academic skills [34-35]. Studies showed that gender bias in the parenting role of children development occurs within the Chinese culture [36-37]. Mothers are often the primary carer for the child, while fathers are less of the carer for the child and work towards providing for the family within the Asian culture . Thus, it is possible that the Chinese father may be less involved in the children’s development than the Chinese mother. However, this study did not examine the gender roles between parent in this study . Furthermore, the researchers reported that the lack of correlation in fathers’ parenting could be due to the idealistic parenting self-report instead of the actual parenting practice.
In Malaysia, Mohamed et al.  conducted a cross-sectional study to examine the influence of family socioeconomic status on children’s social-emotional development. The researchers concluded that family socioeconomic status played a vital role in influencing children’s social-emotional SE development. Similar to the study Ren et al. , this study also reported no significant difference in children’s level of social-emotional development with the father’s income. However, there is a strong relationship between children’s level of social-emotional development with mother’s education level and occupation (i.e., professional, semi-professional, non-professional, unemployed) and the parent’s income status. The findings from Ren et al.  and Mohamed et al.  showed that fathers in the family might have a less significant parenting role within the Asia culture of parenting. The traditional parenting role where mothers are the primary carer for the family might still be rooted in Asia.
Mohamed et al.  conducted a cross-sectional study with 332 early childhood educators in Malaysia. This study found that the early educators in Malaysia had a good overview and knowledge on social-emotional development but lacked knowledge of the factors that influenced the development and how to foster it in the classroom. This study also reported a need to improve and equip teacher’s knowledge of social-emotional development. One study developed and examined the feasibility of a preschool social-emotional competency inventory to screen for children with poor competency . The Preschool Social Emotional Competency Inventory (P-SECI) has a reliability scores between .95 to .98 and was reported valid in predicting children’s social-emotional competencies. Although this study might have high reliability scores, the researchers did not provide information on how the reliability and validity of the P-SECI were achieved. Furthermore, there was no information on whether the participant’s social-emotional competencies were measured before the pilot study. The reliability and validity of the P-SECI remained questionable as the information provided by the researchers was lacking.
In Singapore, an interventional study examined the efficacy of a speech and drama programme on the social-emotional development of children with dyslexia . The children who attended the intervention programme showed improvement in their social-emotional functioning. However, this interventional study did not include a control group. Thus, this study might suffer from low internal validity. Moreover, participants in this study were aware of the purpose of the intervention program. Therefore, the results might have been affected by criterion contamination.
Ong et al.  conducted a 9-year longitudinal study on social-emotional development in Singapore to investigate whether parenting was a moderator between children’s early social-emotional competence and later mental health. The authors concluded that perceived parental care was associated with the early development of social-emotional functioning, while paternal care was especially important for children with more externalising problems. Yeo et al.  reported that children with physical disabilities were comparable in their self-esteem and academic achievements with typically developing children. However, children with physical disabilities experienced peer problems and were likely to participate less in school activities. This study's findings were crucial as there has been scant research on inclusive education in the Asian context.
In Hong Kong, Li et al.  developed and validated the teacher-reported Chinese Inventory of Children’s Socioemotional Competence scale (CICSEC) for Chinese kindergarten children. The CICSEC were reported to have demonstrated good psychometric properties whereby the overall CICSEC and the four subscales had excellent internal consistencies. The criterion validity analysis against the Strength and Difficulties Questionnaires (SDQ)  positively correlated with school readiness and negatively with problem behaviours. The CISEC is one of the culturally relevant social-emotional scales developed in Asia where the items in the questionnaires are consistent with the collectivistic culture in Asia. The exploratory and confirmatory analyses on the CICSEC have found that the children’s social-emotional competence was best represented by a four-factor model that included cognitive control, emotion expressivity, empathy and prosocial behaviours, and emotion regulations across child gender and grade level. This exploratory and confirmatory finding of the four-factor model further confirmed the existing debate that children’s social, emotional, and cognitively skills are interdependent yet distinct [2,17,47]. Lam et al.  examined family characteristics associated with social-emotional development and found that boys had lower social-emotional competence than girls, and older children had a higher level of social-emotional competence. This study also reported that having more siblings positively enhanced social-emotional competence in children. Another study conducted by Lam et al.  was a Social-emotional Well-being of Early Childhood (SEWEC) interventional study that evaluated the social-emotional well-being of participants in an early childhood project within a Hong Kong kindergarten. The intervention programme significantly improved social-emotional competence and reduced anxiety-withdrawal and anger-aggression in children post-intervention.
In Turkey, Seyhan et al.  examined the effects of the Promoting Alternative Thinking Strategies (PATHS) preschool program that measured children’s social-emotional development, the perceived relationship between teacher and children, and teachers’ ability to create a positive classroom environment. This experimental study found that students in the intervention group had significant improvement in social-emotional skills, interpersonal relationship skills, and emotion regulation compared to the control group. Furthermore, students and teachers in the intervention group perceived more positive and dependency in their relationship than those in the control group. However, it is important to note that the observer in this experiment was not blinded to the study, and the findings might therefore be subjected to observer bias. Another Turkish researchers performed a validation study on the Social-Emotion Assessment/Evaluation Measure-Preschool (SEAM) . The confirmatory factor analysis for the Turkish SEAM supported the original factor structure. The researchers concluded that the SEAM measurement was reliable and valid measure for the Turkish population.
In Thailand, Intusoma et al.  reported that educational television viewing was beneficial for children’s social-emotional development. The study reported that 30 to 120 minutes of the educational program per day reduced the risk of poor social-emotional competence relative to non-viewers. However, television viewing for more than 2 hours was reported unhealthy for social-emotional competence. Unfortunately, the researchers did not indicate whether the educational television programme in this study was consistent across all participants. It is unclear which educational television programme was beneficial for the children’s social-emotional competencies. The researchers also categorised documentaries as education programmes due to the narrative nature that is slower in pace. van Driessche et al.  conducted a cross-sectional study in India to assess the predictive factors of caregiver’s burden, psychological comorbidities in the families of children with hearing impairments. This study found that low educational attainment and domestic violence were associated with caregiver’s burden in parents of children with hearing disabilities. Dissatisfaction with family support, behavioural problems in children, and domestic violence were strong predictors for parental psychological morbidities, and they influenced social-emotional development in children with hearing disabilities. However, there was no formal hearing assessment conducted on the participants involved in this study. There is a possibility for children with a mild hearing problem not reported in this study and considered normal hearing.
A cross-sectional study was conducted in Korea to examine the relationship between social-emotional development, gender, age, temperament, and maternal parenting behaviours . Consistent with most studies, this study found that caregivers evaluated boys with more externalising behaviour problems than girls. There was a negative correlation between children’s adaptability and externalising behaviour problems. Kim et al. reported that overprotective or permissive parenting was associated with low social ability in children . On the other hand, refusal or neglect parenting was associated with externalising problem behaviours in children. In Indonesia, an observational study was conducted to examine the effect of an environmental education project approach on children’s social-emotional development . This study found that the approach significantly improved children’s social-emotional development by 22% and increased children’s opportunity to interact with other people.
For this review study, social skills and social competence were coded into one theme, as various studies used different terms to describe children's social competence. Five of the seven studies conducted on social competence used the term social competence, while the two studies in Hong Kong used the term social skills to measure children’s social domain [55-56].
In Hong Kong, Ren et al.  conducted a longitudinal study to examine the antecedents, such as the child’s gender, family socioeconomic status and extra-curricular activities involvement, associated with the children’s social skills. This study found that only children from the lower socioeconomic status participated in the extra-curricular activities had significant improvement in mathematics and reading skills, and not those from higher socioeconomic status. Moreover, this study also reported that parent of children from the higher socioeconomic status were more likely to enrol their children in non-academic extra-curricular activities than those of lower socioeconomic status. This finding highlighted the possible differences in parent socioeconomic background and their preference in extra-curricular activities for their children (i.e., academic versus non-academic). However, Ren et al.  reported no correlation between extra-curricular activities and the development of social skills.
Tong et al.  conducted a mixed-methods study to examine school-wide behaviour interventions implemented in Hong Kong schools and explore teachers’ beliefs about the social skills programs implemented in schools. This study found that most teachers were aware of the benefit of the intervention and support the school-wide behaviour intervention implementation for children with social-emotional behaviour difficulties. The findings in this study were consistent with the study in Malaysia, where teachers were aware of the benefit of social-emotional development. However, they lacked the training to implement the intervention in school . The teacher’s knowledge of social-emotional development was not examined in this study.
Lee et al.  conducted a cross-sectional study and found that both the Korean and United States student in the gifted group has a higher level of social competence than the non-gifted groups. The study also found differences between Korean and United States gifted children in how they rated their social competence ability. The gifted Korean students rated themselves as better at resolving conflicts. In contrast, the gifted United States students rated themselves as better at asserting influence and getting along well with others. Furthermore, Lee et al. reported that female students had a greater ability to make close friends than males. The study finding is in contrast with the others, which reported that gifted children might experience more difficulty creating a relationship with their peers [58-59]. Gifted students in this study could make friends with others, possibly because they recruited samples from the same academic centre. The samples might have the same level of giftedness. It is unclear whether gifted students can establish a relationship with others that are of average intelligence.
Bimla et al.  examined the association between social competency and self-concept with rural children in India. This study found a significant positive relationship between children’s self-concept and social competence. Yoleri  found a significant positive relationship between children’s temperament traits and social competence in Turkey. Yoleri also reported a significant positive relationship between the level of anger and the reactivity/ withdrawal temperament. The findings from both the studies were in line with past research from outside of Asia . Thus, these findings suggested children’s temperament and self-concept might predict their level of social competence despite cultural differences.
In Japan, Anme et al.  carried out a cohort study to describe the Interaction Rating Scale (IRS) features as an evidence-based index of children’s social skills and the quality of parenting. This study found that the IRS could measure children’s social skill development and the quality of parenting with high validity. However, the researchers did not mention how the validity of the IRS was measured. Furthermore, there was no demographic information provided on the clinical population in this study. This study could not be generalized as the validation for IRS was using a cohort sample from the Japan Science and Technology Agency (JST) project. In Korea, Roh et al.  found that children who participated in a 7-week school-based social skill program significantly increased in the degree of their peer relationships. Children from the middle school who did not have peer relationships prior to the 7-week school-based social skills program were reported to establish peer relationships with children of the same age. It is important to note that the effect of the school-based social skills program was measured using the Name Generator Question, where the participants were required to nominate the name of their peers that correspond to each question. A participant might nominate the name of their fellow peer without establishing or forming any relationship. Furthermore, this was a pre-post intervention study. Hence, other factors might have been attributed to the result of this study.
Three out of the five studies on emotional development focused on maternal attitude and children’s emotional development. Two of these studies were from India, and the other study was from Korea. All three studies reported that maternal attitude negatively correlated with children’s emotional development. The two studies from India reported that children’s self-reported emotional dysregulation partially mediated the relationship between the mother’s self-reported non-supportive responses and children’s behavioural problems. Additionally, a factor analysis that compared maternal socialisation behaviour in the Asian context (India and China) found that parental expressive encouragement was unrelated to children’s problem behaviour . However, in a Western context, parental expressive encouragement was a supportive response. Raval et al.  study findings were consistent with the study conducted in Korea . The researchers in Korea reported that children whose maternal controlled positive emotional expressiveness were negatively correlated with behaviour problems positively correlated with social competence development . Thus, these findings highlighted the norm differences in parental emotion expression and the importance of understanding children’s functioning in a cultural context.
A study from Pakistan compared the levels of adaptive emotional abilities between adolescents with hearing disability and normal hearing adolescents . This study also examined the sociodemographic variables that might predict the emotional development of adolescents with hearing disability. Consistent with past studies, this study found that adolescents with hearing disability scored significantly lower on the adaptive emotional abilities scale than normal hearing adolescents. Akram et al.  explained that hearing disability itself was not the only cause that led to poor adaptive emotional ability. The sociodemographic variables showed that the accessibility and availability of hearing and speech services, the presence of hearing-impaired family members, the preferred communication language between the adolescent and family members were associated with the adolescents’ adaptive emotional abilities.
In Taiwan, Chang et al.  carried out a study that assessed the level of emotional development among Taiwanese children based on Dabrowski’s theory . The researchers also assessed whether emotional development and over-excitability predicted personal adjustment in gifted students and normal students. There were 123 mathematically gifted students and 132 normal students aged 16 to 18 years old involved in this study. Based on Dabrowski’s theory , emotional over-excitability is the most important characteristic that effectively predicted the level of emotional development among gifted students. However, this study found a negative correlation between emotional development and emotional over-excitability. The researchers reported that cultural-specific variables might explain the discrepancy found in this study. Given the discrepancy in findings, it is important to note that this Taiwanese study only included mathematically gifted students in their study, and the findings did not represent other gifted students in Taiwan.
Social-Emotional Learning (SEL)
Wenling et al.  carried out a cross-sectional study with 375 early childhood educators to examine teachers’ perceptions of SEL in early childhood centres in China. This study found that teachers’ perceptions of SEL in China were at a moderate level. In addition, female teachers with higher qualification and teaching in private schools were reported to be more supportive of SEL than male or female teachers with lower qualifications or those taught at public schools. These findings were consistent with the Malaysian study that examined teachers’ perceptions of SE development in children . Furthermore, the Malaysian teachers had a lower understanding of the factors associated with social-emotional development in children and reported that they did not know how social-emotional skills could be taught in a classroom setting.
One of the three intervention studies on SEL, Pinchumphonsan et al.  reported that the 15-week intervention program had significantly reduced the number of problem behaviours and improved positive behaviours in children after completing the program. However, similar to the other pre-post intervention studies mentioned above, the improvement in participants’ behaviours could be attributed to other factors unrelated to the intervention program. Iaosanurak et al.  developed an SEL intervention program to compare Thailand and Cambodian cultural groups and gender differences. However, gender differences were the only differences found in the SEL intervention. The level of empathy and responsibility scores in female participants increased after the intervention program. The cultures in these two countries may be highly similar, which led to a non-significant result when compared across cultures. In Hong Kong, Lam et al.  implemented and evaluated the effectiveness of a mindfulness program among four of the lowest academic tier groups in public schools. This study found significant differences between the intervention and control groups with medium to large effect size on emotional control, working memory, self-monitoring and anxiety/depression.
Problem behaviour and prosocial behaviour were coded into one domain to review these studies systematically. Yang et al.  conducted a cross-sectional study in China to estimate the prevalence of behavioural problems and their risk factors among school children aged 6 to 16 years old in Beijing. There were 9,295 students from urban and suburban districts in Beijing who participated in this study. The Child Behaviour Checklist was used as a screener to screen for emotional and behavioural problems. The detection rate of behavioural problems in this study was 16.7%. However, the rate of behavioural problems decreased with age. The researchers reported that behavioural problems were more significant in children aged 6 – 11 years old and not significant in children aged 12 – 16 years. The researchers found that girls experienced more internalising behavioural problems (e.g., depression, withdrawal, anxiety), and boys experienced more externalising behavioural problems (e.g., aggressive behaviour, social problem, hyperactivity). These researchers also found that older children had a better level of social competence. Yang et al.  explained that the behavioural problems in the younger age groups might be due to China's one-child policy. These children did not have siblings and might have fewer opportunities to interact with other children outside school.
In China, Guo et al.  found prosocial behaviour as a predictive factor for academic success, and that peer acceptance mediated between prosocial behaviour and academic success. These findings were consistent with the literature in the Western cultural context . In Indonesia, Dewi et al.  conducted quasi-experimental research to assess the effect of the traditional game (Magoak-goakan) on developing prosocial behaviour in preschool children in Bali. Fifty-two preschool children aged 5-6 years old were involved in this placebo control study (1:1 ratio). This study reported a significant increase in the intervention group's prosocial behaviour while no changes were found in the control group. Although this is a small sample size study, it is worth noting that situational factors such as the traditional game of Magoak-goakan might influence the development of prosocial behaviour. More research is required to examine the situational factors in children’s prosocial behaviour within the Asia context.
Sun et al.  examined the development of cool and hot self-regulation in a Chinese sample of pre-schoolers in Hong Kong. This study also examined the relationship between cool and hot self-regulation with children’s academic achievement, behavioural problems, general knowledge, and fine motor and gross motor skills. This study found that the cool self-regulation domain positively predicts children’s academic achievements, general knowledge, and fine motor and gross motor skills. In contrast, the hot self-regulation domain positively predicted children’s gross motor skills only. Both cool and hot self-regulation negatively predicted children’s hyperactivity.
Sun and Kang  reported that cool and hot self-regulation might have a distinct structure among Chinese pre-schoolers. Sun and Kang could not find the exploratory factor analysis and confirmatory factor analysis fit that showed the relationship between cool and hot self-regulation was similar to those found in Western samples. The researchers concluded that the Chinese culture differed from Western culture, where the Chinese culture emphasised children's behavioural conformity, obeying adults, and inhibiting inner impulses. The cultural differences found in this study might then explain why most of the children, as young as three years old in this study, could regulate their behaviours despite the emotional or motivational triggers in the hot self-regulation experimental task.
Zhi et al.  explored the relationship between children’s self-control and family savings for children’s future education using the China Family Panel Studies data. This study found that children who lived in families with savings for their future education had higher self-control than families without savings. The effect size for this study was small (d = .06).