1 Introduction

High Abilities/Giftedness (HA/G) is characterized by the high potential of aptitudes, talents and skills, evidenced in the high performance in the various areas of human activities including academic, demonstrated since childhood. Such areas include the intellectual, academic, leadership, psychomotor and arts fields, among others (National Policy for Special Education from the Perspective of Inclusive Education [PNEEPEI], 2008; Law no. 12,796, April 4, 2013).

The identification of children with HA/G has been based on the theoretical perspective of the Three Ring Model (^{Renzulli, 1998}, 2011). In this conception, HA/G is due to the confluence of three factors: above average intellectual ability, task involvement and creativity. The current definition of HA/G proposed in the PNEEPEI (2008) document is strongly influenced by this model. The HA/G is covered by the Special Education guidelines (Law no. 9,394, of December 20, 1996; Decree no. 7,611, of November 17, 2011), with the aim of guaranteeing children access to specialized services. This means that it is the school’s role to prepare to meet their needs and particularities.

Regarding inclusion policies, it is understood that interpersonal development (specifically empathy skills and prosocial behaviors) is an indispensable component of this inclusive process. The goal is to improve the quality of social relationships by promoting attitudes of understanding and acceptance of differences by peers, teachers and school staff. Considering, then, the importance of acquiring and maintaining these skillful behaviors, the school needs to invest in the interpersonal development of students, and even of teachers, as they mediate classroom relationships and conflicts (^{Del Prette & Del Prette, 2013}).

^{Castro and Bolsoni-Silva (2008}) state that, in fulfilling this role, teachers can maintain, strengthen or even discourage behaviors related to child-child and child-teacher interaction, influencing both academic and social aspects. According to ^{Bandeira and Quaglia (2006}), to be successful, children need to have assertive behaviors, as this contributes to improve interpersonal communication, expressing feelings and needs. However, it is up to the teacher to identify such behaviors, seeking to stimulate them in different situations.

From this perspective, it is undeniable that educational systems should equally prioritize the development of social skills of all their students, whether or not this target population is from the Special Education area. From the beginning of schooling, the educational environment should favor and enhance the social and academic development of students in view of an adequate formation for life in all its unfolding.

Assuming that child behavior is the result of the child’s behavioral history in the environments in which he/she lives and is also maintained by family environmental contingencies, in addition to teachers, it is essential to work with parents and/or guardians (^{Silvares, 1993}). In this sense, ^{Bolsoni-Silva and Loureiro (2011}) compared the parenting practices and child behavior of two groups of children, clinical and non-clinical, for behavioral problems. The results showed that the behaviors that differentiated the groups were, above all, those related to positive parenting practices and children’s social skills, confirming the influence of family environmental contingencies pointed out by ^{Silvares (1993)}. Family support acts as a protective factor and a promoter of resilience and with children with HA/G is no different, as it is fundamental for the development of talent (^{Chagas, 2008}; ^{Chagas & Fleith, 2009}; ^{Dessen, 2007}).

There are authors who suggest that talented young people, especially those with extreme abilities, are more vulnerable to emotional and social problems and higher risks for depression, anxiety and suicide (^{Alencar, 2007}; ^{Antipoff & Campos, 2010}; ^{Cassady & Cross, 2006}; ^{Chagas & Fleith, 2010}; ^{Mirnics, Kovia, & Bagdy, 2015}; ^{Peterson, 2009}). The development of their skills, quality of life, success or self-realization therefore depends on a complex set of individual and environmental variables (^{Chagas & Fleith, 2010}). Therefore, an intervention focused on this area can help to prevent or reduce these deficits and mismatches (^{Matos & Maciel, 2016}) and ideally this should be done as early as possible.

^{Chagas and Fleith (2010}) have highlighted the need for broadly planned and employed activities aimed at the development of social skills in schools, especially in programs that serve this audience. To the authors, the activities should be systematically planned, aiming to help: in the strengthening and construction of positive self-image, self-esteem and selfconcept; in building an integrated identity; in developing nonviolent bullying strategies; and maintaining a positive interpersonal relationship without, however, relinquishing their skills and interests or even denying talent.

According to ^{Del Prette and Del Prette (2013}, ²⁰¹⁷⁾, social skills are behaviors that, in a given situational-cultural context, are highly likely to produce reinforcers for the individual and minimize aversive stimulation, contributing to the quality and effectiveness of the relationship with others. Therefore, they constitute a specific class of behaviors that an individual issues to successfully complete a social task, such as joining a group of classmates, making friends, initiating and maintaining conversation, among others.

These skills can be implemented or developed in Social Skills Training (SST) programs, which correspond to the set of planned activities that structure learning processes for social behaviors, mediated and conducted by a therapist or coordinator. In this type of program, the goal is to increase the frequency and/or improve the proficiency of already learned but impaired social skills, and to teach new and significant social skills by diminishing or extinguishing competing behaviors with such skills (^{Del Prette & Del Prette, 2017}).

Among the characteristics of people with HA/G, there are contradictions in the literature regarding the presence or absence of social skills. Some studies indicate that children with HA/G have deficits in social and emotional areas, as mentioned earlier, while other studies indicate that this population has a good repertoire of social and emotional skills (^{Bain & Bell, 2004}; ^{França-Freitas, Del Prette, & Del Prette, 2014}, ²⁰¹⁷; ^{Freitas & Del Prette, 2013}; ^{Galloway & Porath, 1997}; ^{Loos-Sant’Ana & Trancoso, 2014}; ^{Prieto, Ferrándiz, Ferrando, Sánchez, & Bermejo, 2016}; ^{Versteynen, 2001}).

In research conducted in national scientific databases with the keywords “social skills”, “high abilities”, “giftedness”, “high abilities or giftedness” and “talent”, no studies of social skill interventions with students with HA/G, with parents or teachers were found. The studies found describe the social skills of children with HA/G from their own reports (^{França-Freitas, Del Prette, & Del Prette, 2014}; ^{Loos-Sant’Ana & Trancoso, 2014}) and according to the evaluation of the teachers (^{Freitas & Del Prette, 2013}). In the international literature, studies relating the two themes were identified: social skills interventions with children with HA/G (^{Calero-García & García-Martín, 2005}; ^{Gómez-Pérez et al., 2014}) and intervention with parents of children with HA/G on parental skills (^{Morawska & Sanders, 2009}). Thus, this work presents unprecedented data as it was worked with three groups of participants: students, parents and teachers.

^{Calero-García and García-Martín (2005}) performed an intervention with nine children, between 6 and 10 years old, with intellectual giftedness who, according to their parents, presented relationship problems with their peers and characteristics of introversion, distancing, critical posture and coldness. There were 10 sessions of 60 minutes each, in which the following topics were addressed: communication techniques; identification of situations of interpersonal problems; perceptions of feelings of others; empathy; ability to respond to failure; consideration of all factors present in situations of social interaction; analysis of conflict resolution options; delimitation of the consequences of actions; and planning and decision making. The results showed that children’s interpersonal skills improved significantly after the intervention.

^{Gómez-Pérez et al. (2014}) evaluated the effectiveness of an interpersonal problem solving skills program conducted with 40 gifted children aged 7 to 13 years old divided into experimental and control groups. According to parents, children had interpersonal, behavioral and peer problems, and emotional difficulties. The program was conducted in groups in 10 sessions, using the same themes of the study conducted by ^{Calero-García and García-Martín (2005}). According to the authors, after the program, the interpersonal problem-solving skills of children in the experimental group increased, demonstrating the positive effects of the program.

^{Morawska and Sanders (2009}) evaluated the effects of a behavioral intervention with 75 parents (divided into experimental and control group) of children aged 3 to 10 years old with HA/G. The purpose of the intervention was to improve parenting skills and to evaluate the effect of these changes on children’s social and emotional behavior. The program, developed in nine meetings, involved management skills in the areas of child development promotion, misbehavior management, and planned activities and routines to improve the generalization and maintenance of parenting skills. The results indicated significant effects of the intervention for parents of the experimental group who reported behavioral problems and hyperactivity of their children, but had no effects on the emotional environment. Parents also pointed to significant improvements in their own parenting style. The follow up, after six months, pointed to the maintenance of the positive effects of the intervention. However, according to the teachers, there were no changes in children’s behavior, except for hyperactivity. However, they had evaluated children with few behavioral problems before the intervention.

Although the subjects were not children with HA/G, ^{Pereira (2013}) coordinated a study with seven municipal schools in a city in the state of Mato Grosso do Sul, Brazil, with 28 teachers and approximately 700 Elementary School students, aiming to develop social skills and prevent behavior problems. Each teacher indicated up to three students with behavioral problems and three with socially accepted behaviors, who were academically evaluated. Twenty 50-minute sessions were held weekly with each class and their teachers. The themes of the intervention were: socialization, communication, expression of feelings, knowing how to defend oneself and collaboration. After the intervention, teachers reported positive points of the intervention, such as: rule making, improvement in the development of group activities and more frequent student participation. Regarding the academic performance of previously evaluated students, there was improvement after the intervention for both groups.

According to ^{Del Prette and Del Prette (2013}), comparative studies between different data sources, such as parents and teachers, can broaden the categorization and understanding of the social skills repertoire, helping to identify these behaviors in children with HA/G in different social contexts, such as school and family. The authors advocate investing in SST programs in two strands: the first as an alternative to prevention through integrated action between school and family; and the second associated with clinical interventions with the child, aimed at overcoming interpersonal difficulties and associated problems. In this work, the SST will be done considering the first strand.

Studies on the repertoire of social skills of children with HA/G are inconclusive, which justifies studies that investigate this theme with this population, in order to collect information from multiple informants: themselves; their parents and teachers. Also, there were few studies found related to interventions with this population as well as significant adults from their immediate contexts, such as family and school. It is important to test effectiveness (assessing whether or not the proposed objectives have been achieved) and efficiency (quick and least cost-effective procedure) of interventions with the child, parents and teachers, considering that they make up contexts of face-to-face interaction and extended time, which can mean a consistent improvement in the child’s development environment.

This study aimed to describe and compare the repertoire of social skills, behavioral problems and academic competence of students with HA/G: 1) according to their own account, before and after a program on social skills; and 2) according to the report of their parents/ guardians and teachers, before and after guidance on HA/G and social skills.

2 Method

Nine students with HA/G participated in ^{Mendonça’s (2015}) study, who attended the 1st, 3rd, 4th and 5th grades of Elementary School of the São Paulo state school, as well as their respective parents/guardians and teachers. Among the students, there were six boys and three girls aged between 6 and 10 (average = 8.44 and SD = 1.42). The nine students were divided into two groups⁶ according to the school shift: Group 1 (G1), morning students of the 3rd, 4th and 5th grades composed of three boys and two girls; and Group 2 (G2), students of the afternoon of the 1st and 3^rd grades composed of three boys and one girl.

Regarding the teachers, the eight were aged between 28 and 63 years old (average = 43; SD = 11.43) and with different levels of education: teacher-training graduate (1), university undergraduates (5) and graduates (2), who worked in regular class. They were divided into two groups according to the students under their responsibility: Group 1 (G1), four teachers of the morning period; and Group 2 (G2), four teachers of the afternoon period.

Among those responsible for the child, five were mothers, one grandmother and two fathers, aged between 30 and 56 (mean = 39.75; SD = 9.93). Regarding the academic level, one mother had completed Elementary School, five with completed High School and two with completed Higher Education. The parents/guardians were divided into two groups according to their children’s groups: Group 1 (G1), five parents/guardians of the morning students; and Group 2 (G2), three parents/guardians of the afternoon students. The person responsible for student 4 did not participate in the intervention.

3 Results and discussion

The data presented in Figure 1 show, for each Student (S), the percentiles of each factor of the SSRS-BR social skills scale according to themselves in the pre, posttest and follow-up.

Figure 1 Student Performance (S) in the three assessments (pre, posttest and follow-up) in the SSRS-BR, answered by them.Note - Social skills percentile: 0-25 = Below average repertoire; 26-35 = Average repertoire; 36-65 = Good repertoire; 66-75 = Elaborated repertoire; 76-100 = Highly elaborated repertoire. 

In the pretest, eight of the nine participants had deficits (0-25 percentile) in some of the social skills factors (S1, S2, S3, S4, S5, S7, S8, S9). This data contradicts the findings of ^{Loos-Sant’Ana and Trancoso (2014}), whose minority of participants reported having belowaverage social repertoire. Still in the pretest, all presented deficits in at least three social skills factors, different from the data found by ^{França-Freitas, Del Prette and Del Prette (2014}). Their participants presented more elaborate repertoires of social skills in the factors of responsibility, assertiveness, self-control, civility and expression of positive feeling.

In the posttest that corresponds to the evaluation following the intervention, the frequency of percentile skills increased or remained in all factors for students S1, S3, S4 and S5. Six presented deficits (S1, S3, S4, S7, S8, S9) in some ability and, in the follow-up, three presented deficits in some ability (S1, S2, S9). For the other students the frequencies varied. The position of the behavioral percentile that corresponded to good to highly elaborated repertoire (36 to 100 percentile) in the posttest were: in general social skills for S2, S3, S4, S5 and S6; in Empathy/affectivity for S2, S3, S5 and S6; in Responsibility for S1, S2, S3, S5, S6 and S8; Self-control/civility for S2, S5 and S6; and in Assertiveness for S2, S3, S4, S5 and S6.

In the follow-up, it is observed that the position of the percentile of behaviors that corresponds to good to highly elaborated repertoire (36 to 100 percentile) were in general social skills for S3, S4, S5, S6, S7, S8 and S9. Empathy/affectivity for S3, S4, S5, S6, S7, S8 and S9. Responsibility for all students. Self-control/civility increased for students S3, S4, S5, S6, S7 and S8. Assertiveness increased for S2, S3, S4, S5, S6, S7, S8, and S9.

It can be considered that the intervention was relevant for some participants considering the posttest. However, considering the follow-up, some results were not maintained. With the exception of S1 and S2, it was pertinent for all participants, since the frequency of social skills increased and, consequently, the percentile position as well. This means that, after the intervention program, students presented between good and highly elaborated repertoire in social skills according to self-report, not staying at the same level for two of the participants in the follow-up.

As students had very low pretest scores, it is possible that the results observed in the posttest and follow-up may be due to the intervention with them and possibly with their teachers. According to ^{França-Freitas, Del Prette and Del Prette (2014}), responsibility was one of the most scored skills by students with HA/G. In this research, this was the ability that students reported to be more frequent in the pretest and also the one that had the most gains with the intervention, observed in the posttest and follow-up. This result is in line with the studies conducted by ^{Bain and Bell (2004}) and ^{Chagas (2008}), in which gifted and talented adolescents reported a higher frequency of behaviors indicative of responsibility in performing activities.

Figure 2 shows the percentile position on each factor of the students’ SSRS-BR social skills scale and behavioral problems according to their parents/guardians (P/G) in the pre, posttest, and follow-up.

Figure 2 Student Performance on Social Skills (SS) and Behavioral Problems (BP) in the three SSRS-BR assessments, answered by parents/guardians (P/G).Note - SS Percentile: 0-25 = Below average repertoire; 26-35 = Average repertoire; 36-65 = Good repertoire; 66-75 = Elaborated repertoire; 76-100 = Highly elaborated repertoire. BP percentile: 0-25 = Too low repertoire; 26-35 = Low repertoire; 36-65 = Median repertoire; 66-75 = Average repertoire; 76-100 = Above average repertoire. 

Four students presented pretest deficits in some of the social ability factors (S1, S2, S7, S8), five presented deficits in the posttest (S1, S2, S3, S7, S8) and, in the follow-up, seven (S1, S2, S3, S5, S6, S7, S8). One student (S7) presented behavior problem in the pretest, considered as needing intervention. In the posttest, no student presented behavioral problems requiring intervention and, in the follow-up, one student (S3) presented it. One hypothesis for these results is that some parents/guardians became more critical of their children’s behavior after the intervention, taking as a model what or how they would like their children to behave.

Considering behavioral problems, on the one hand, internalizers increased in the posttest and decreased in the follow-up, however, the score was still higher than in the pretest. On the other hand, according to the teachers, all student behavior problems decreased in the posttest, both internalizing and externalizing, as observed in Figure 3 further below.

Figure 3 Student Performance on Social Skills (SS), Academic Competency, and Behavioral Problems (BP) in the three SSRS-BR assessments, answered by the Teachers (T).Note - SS Percentile: 0-25 = Below average repertoire; 26-35 = Average repertoire; 36-65 = Good repertoire; 66-75 = Elaborated repertoire; 76-100 = Highly elaborated repertoire. BP percentile: 0-25 = Too low repertoire; 26-35 = Low repertoire; 36-65 = Median repertoire; 66-75 = Average repertoire; 76-100 = Above average repertoire. Academic competence percentile: 0-25 = Below average; 26-35 = Lower average; 36-65 = Median; 66-75 = High; 76-100 = Very high. 

Figure 3 shows the percentile position in each factor of the students’ SSRS-BR social skills, academic competence, and behavioral problems according to their teachers (T) in the pre and posttest. Three students had pretest deficits in some of the social ability factors (S1, S3, S7), and three had pretest behavioral problems considered in need of intervention (S6, S7, S9). In the posttest, four presented deficits (S1, S4, S6, S9) and three presented behavioral problems requiring intervention (S3, S4, S6).

The teachers did pre and posttest evaluations only. The abilities with the highest posttest scores, in order of improvement, were: cooperation/affection, social assertiveness/social resourcefulness, general social skills, and self-control. Academic competence also increased posttest. Such improvements can be attributed to the intervention and also to the improved behaviors of the students who were under intervention.

Interestingly, for most participants - students, parents/guardians, and teachers - there were improvements in skillful behaviors and fewer problems in the posttest and follow-up. One hypothesis is that perhaps parents/guardians and teachers have changed their perception of students, and this may have been responsible for the improvement. Another possibility is habituation to the instrument, but after 10 and 12 months this possibility may be ruled out.

From the objectives of this study, it is observed that, in the pretest, students rated themselves worse than their parents/guardians and teachers assessed them. Also, it was observed that their parents/guardians rated them better in the pretest than in the posttest and at the follow-up. According to ^{Skinner (1974}), it is common to report that participants become more critical after an intervention.

The teachers were the ones that reported more social skills in the pretest and posttest, reporting more often than the parents/guardians, and especially more than the students themselves. The findings of ^{Morawska and Sanders (2009}) and ^{Bolsoni-Silva and Loureiro (2016}) also corroborate the fact that female teachers reported more student skills than the parents.

Considering behavioral problems, parents/guardians reported more general and externalizing behavioral problems than teachers, both in the pretest and posttest. This data corroborates the findings of ^{Bandeira, Rocha, Freitas, Del Prette and Del Prette (2006}) and ^{Morawska and Sanders (2009}) who pointed out that parents described more behavioral problems than teachers. Internalizing behaviors were more noted by teachers in the pretest and by parents/guardians in the posttest. To parents/guardians, externalizing factors decreased over time, less frequently in follow-up.

4 Final considerations

This study aimed to evaluate the repertoire of social skills, behavioral problems and academic competence of students with HA/G, before and after an intervention program on social skills, developed with the three segments. It was used a universal intervention, focused on age-related issues and common problems present in the relationships between the three segments. Changes were observed, especially in the students’ own reports; in the follow-up, to parents/guardians there was an improvement in responsibility and to teachers in general social skills, self-control, social assertiveness/social resourcefulness, cooperation/affection and academic competence, having the research thus achieved its objectives. A strong point of the research was to characterize the repertoire of social skills and behavioral problems of students with HA/G using, in addition to self-report, the report of their parents/guardians and teachers.

One of the limitations of the study was the small number of participants, which, although difficult to generalize the results, allows the survey and confirmation of new hypotheses. Investigating HA/G and social skills is still considered a challenge for researchers, with several gaps to be filled and deepened. In this sense, this study could contribute to the area as it points out that students with HA/G reported having social skills, but had deficits in some areas, which was also reported by their parents/guardians, justifying the intervention performed. A suggestion for future studies is, from initial data, to propose programs focused on social skills, thinking of practices that minimize problems and expand the skillful repertoire. The teacher’s participation proved indispensable in the process, as they need strategies to promote inclusive environments for this population.

Importantly, in PNEEPEI (2008), Specialized Educational Assistance (SEA) is foreseen for this population, intervention programs of this nature could be developed in the context of the articulation of Special Education with the other teachers of the common room, aiming at the development of general and/or specific skills, as needed and individual demands of each student.

The results indicated that, with the intervention, it was possible to improve the skillful repertoire of some participants in some factors. In terms of implications for interventions with teachers and parents/guardians, it can be pointed out the need for more encounters, both with the theme of HA/G, as with social skills.

Finally, in light of these findings, some suggestions for future research are listed to deepen the understanding of social skills in this population: 1) Assessment of children’s social skills repertoire by their peers (sociometric assessment); 2) Evaluation of social competence through research using observational methods of social behavior, such as self-records, cursive registration and video recording; and 3) Characterization of the repertoire of social skills and behavior problems in individuals with HA/G in another age group, such as adolescents and adults.