Introduction

Student engagement with school has received significant research and educational policy attention given its documented impact on student learning outcomes. Namely, students who are highly engaged in school have higher attendance and achievement than disengaged students (Appleton et al. 2008; Fredricks et al. 2004). Among different facets of school engagement, the degree to which children have positive attitudes toward school has been suggested as a meaningful indicator of emotional or affective engagement with school for elementary school children (Ladd and Price 1987). For example, elementary children’s positive school sentiment is related to participation in learning and achievement (Ladd et al. 2000; Ladd and Dinella 2009). Thus, in an effort to foster children’s learning and academic success, it is important to understand what contributes to school attitudes among elementary school children.

The purpose of this study was to examine the manner in which children’s beliefs about the importance of social skills (i.e., the degree to which children value socially skillful behaviors) contribute to children’s positive school attitudes. Specifically, because beliefs and values guide behavior, we postulated that social behavior (e.g., aggression, social skills) is an important mechanism by which children’s beliefs about social skills are related to school attitudes. Also, research suggests gender differences in a number of social and behavioral characteristics (Zakriski et al. 2005); thus, building upon the literature, we examined gender effects in the study constructs.

Precursors of School Attitudes

Broadly speaking, research on school engagement has identified a number of precursors of student engagement with school among middle and high school students, in particular. They include school characteristics (e.g., school size, clear and consistent goals), classroom structure (e.g., expectations, behavior management), task characteristics (e.g., novelty, opportunities for collaboration), teacher support, and peer affiliations (see Fredricks et al. 2004). Given that school engagement is multi-faceted, including behavioral, emotional, and cognitive engagement, it might be that the most relevant precursors of school engagement further depend on the form of school engagement. For example, task characteristics might be particularly relevant to cognitive engagement whereas teacher support might be particularly relevant to behavioral and emotional engagement (Fredricks et al. 2004).

Social relationships and behavioral characteristics have been identified as important precursors to school attitudes among elementary aged children. For example, children who have positive peer relationships, as indicated by peer acceptance, friendships, and perceived peer support, tend to feel more positively toward school (Boulton et al. 2011). Also, as compared to less socially skilled children, those who are more socially skilled reported greater positive school attitudes (Kwon et al. 2012). In contrast, children with elevated levels of aggression reported decreased positive school attitudes over time (Gest et al. 2005).

Theoretical Framework

The current study was guided by a social cognitive developmental perspective (Crick and Dodge 1994; Olson and Dweck 2008) which suggests that developmental contexts (e.g., parenting practices, peer relationships, cultural context) shape children’s social cognitions such as goals, beliefs, and attributions, which in turn contribute to children’s social, psychological, and academic adjustment. In the current study, we focused on children’s beliefs about social skills as an important social cognition that contributes to behaviors (social skills and aggression) and attitude toward school.

Social cognition plays a significant role in behavior. In the case of aggressive behavior, for example, a hostile attribution bias refers to an individual’s inclination to perceive hostile intent in the face of an ambiguous provocation (Dodge and Frame 1982). Compared to nonaggressive children, aggressive children have been found more likely to perceive hostile intention, experience more anger, and approve hostile reactions in response to ambiguous situations (Graham et al. 1992). A normative belief refers to an individual’s cognition about the appropriateness or inappropriateness of a behavior (Huesmann and Guerra 1997). A normative belief conceivably filters behaviors suggested by social scripts such that an individual’s selection of a behavioral response depends on his or her beliefs about whether or not the behavior is appropriate (Huesmann and Guerra 1997). In fact, a number of studies have shown that children who believe it is acceptable or legitimate to use aggression in response to peer provocation are more likely to display aggressive behaviors (Bellmore et al. 2005; Erdley and Asher 1998; Huesmann and Guerra 1997).

In contrast to research on aggression-related social cognitive factors, little is known about social cognitions related to adaptive behaviors, including children’s beliefs about the importance of social skills. This appears to be a significant gap given research that highlights competence-based behaviors, including social skills, as providing children with important foundations for positive adjustment and school performance (Durlak et al. 2011). For example, as compared to aggression, prosocial behaviors (e.g., cooperation, sharing, helping) have a stronger predictive effect on long-term academic achievement and successful peer relationships (Caprara et al. 2000). Likewise, children’s beliefs about social skills might serve as an important social cognition that contributes to children’s functioning.

Social and Developmental Context of Social Skills

Social skills involve a number of prosocial behaviors such as helping, sharing, and caring, and those behaviors are essentially situated in a social and interpersonal context. For example, a previous study (Youniss 1980) found that children conceptualize kindness differently depending on whether it is directed toward adults or peers. That is, kindness directed toward adults had more to do with compliance and cooperation whereas that directed toward peers included behaviors that promote positive relationships such as empathy and friendliness. Similarly, children’s conceptions of social skills might change as their social context changes with development.

Peers emerge as an important social and interpersonal context for school-aged children. To develop and maintain positive interpersonal relationships with peers, children need to display skillful behaviors that are appropriate and consistent with the expectations in the peer group. For example, Greener and Crick (1999) asked children in third through sixth grade to describe what they believed to be normative prosocial behaviors in their peer group. Participating children identified relationally inclusive acts, such as making friends and including others in group activities as salient types of prosocial behaviors. It might not necessarily be that children do not value traditionally defined prosocial behaviors such as helping and sharing; rather, it appears that relationally inclusive acts emerge as a more salient type of prosocial behavior for children in middle childhood (Greener and Crick 1999).

The importance of considering the developmental and social context of a behavior has been bolstered by a contextual approach to social skills assessment. Given that social demands differ across settings and situations, a contextual approach to the assessment of social skills emphasizes the need to identify skillful behaviors that are congruent with the demands of a specific social context (Sheridan et al. 1999). For example, the social context of the family is different from that of the peer group, and children need to meet the behavioral expectations in a given context to achieve successful social interactions. Based on a contextual approach to social skills assessment, previous studies have identified socially skillful behaviors that are important in children’s peer group (Kwon et al. 2012; Warnes et al. 2005). Likewise, we focused on skillful behaviors particularly relevant to children’s peer group in our conceptualization of social skills and children’s beliefs about them.

Gender Effects

One of the intuitive approaches to studying gender effects in social and psychological attributes is to examine mean level differences in the constructs of interest. For example, boys have been noted to display higher levels of overt aggression than girls, whereas the opposite pattern has been noted for prosocial behaviors (Zakriski et al. 2005). However, gender differences do not seem as evident when the associations among constructs are examined. Bellmore et al. (2005) tested a mediation model assessing whether individuals who believed it was appropriate to use aggression were more likely to choose hostile responses to peer provocation, which in turn led to heightened levels of aggressive reputation among peers. Importantly, findings indicated that the mediation model was similar for both males and female young adolescents. Another study that involved elementary students found no gender difference in the association between social skills and adjustment outcomes; social skills were positively associated with social status (i.e., peer acceptance and popularity) and school adjustment equally for boys and girls (Kwon et al. 2012).

Summary of Present Study

This study had three goals. The first was to examine the effect of children’s beliefs about the importance of social skills on their social behaviors (i.e., social skills, aggression) and school attitudes. The second goal was to examine the indirect effect of beliefs about the importance of social skills on school attitudes through social skills and aggression. We hypothesized that (a) beliefs about the importance of social skills are positively related to social skills, (b) beliefs about the importance of social skills are negatively related to aggression, (c) social skills are positively related to positive school attitudes, (d) aggression is negatively related to positive school attitudes, and (e) beliefs about the importance of social skills are indirectly related to positive school attitudes via social skills and aggression. The third goal was to examine the gender effect in terms of the mean levels of and the associations among study variables. We hypothesized the presence of gender differences in the mean levels of the study variables (e.g., higher levels of social skills for girls than for boys and higher levels of aggression for boys than for girls) but not in the associations among the variables (e.g., beliefs about the importance of social skills are positively related to social skills for both boys and girls).

Methods

Participants

A total of 342 (180 male, 162 female; mean age = 9.70 years) students participated in this study. Students were from three elementary schools in Midwestern rural communities of the United States and were enrolled in grades 3 (n = 112), 4 (n = 142), and 5 (n = 88). School records indicated that 94 % of participating children were White.

Participating students’ classroom teachers (N = 22) also participated in this study. The majority of teachers (86 %) were female, and all self-reported as White. Their average years of teaching was 17.82 (SD = 10.66).

Procedures

Study procedures were approved by the Internal Review Board of the large research institution at which the study was conducted. For recruitment, parents were sent a letter describing the study with a form requesting informed consent for their child’s participation. Teachers also received a letter with a consent form for their participation. This study involved a peer nomination procedure; we used a grade-based, as opposed to classroom-based, peer nomination procedure because it was determined that the former reflected the nature of participating children’s peer interactions more accurately in the rural communities than the latter (e.g., Kwon et al. 2012). That is, participating schools were recruited from rural areas, and school personnel explained that children were highly familiar with their peers across classrooms through frequent interactions in and out of school. There were two to five classrooms per grade, and we determined consent rate across classrooms within a grade level (e.g., 3rd graders at school 1). For a particular grade level to participate, at least 75 % of students in the grade were required to give consent to ensure validity of peer nomination measures (see Hamilton et al. 2000). A total of four third grades, four fourth grades, and three fifth grades were initially recruited. Of these, two third grades, two fourth grades, and two fifth grades met the required consent rate; the consent rate of participating grade units ranged from 77 to 91 %.

The child questionnaires were administered to participants in classrooms for approximately an hour. Prior to the questionnaire administration, child assent was obtained. Children were informed that they were allowed to stop participating in the study at any time. One research team member read aloud the questionnaire to students in the classroom while the other member provided individual assistance as necessary. For peer nomination measures, which were part of the child questionnaire, children were provided a grade level roster which included the names of students whose parents provided consent. A unique number was assigned to each participant, and students were to use the number identifier, instead of name, for the peer nomination items. Non-participating children were asked to read or draw quietly at their desks. Regardless of participation, all students in a participating grade received stationary gifts. Teachers completed the ratings for each participating child at their convenience. They returned the questionnaire within a week after distribution, and they were provided a monetary honorarium for their time.

Measures

Beliefs About the Importance of Social Skills

The Beliefs about the Importance of Social Skills scale assessed the degree to which children value a range of socially skillful behaviors pertaining to peer interactions. Socially skillful behaviors in children’s peer group were identified in previous research (Kwon et al. 2012; Warnes et al. 2005). Specifically, children were instructed to think about times when they play or work with other children and how important they think different social behaviors are in their interactions. Example items included: “to offer to help other kids when they need it,” “to show other kids that I care when they are sad,” “to invite other kids to do things together,” and “to accept other kids who are different from me.” Children rated each of the 25 items on a four-point scale that ranged from 1 (not important) to 4 (very important). The alpha of the Beliefs about the Importance of Social Skills scale was .88.

Social Skills

Children’s social skills were assessed by peers based on a peer nomination procedure. Children nominated up to three peers in their grade who fit each of the descriptions of social skills (e.g., “this person shows other kids that he or she cares when they are sad,” “this person accepts other kids who are different from him or her.”). The same 25 items from the Beliefs about the Importance of Social Skills scale were used in the peer nomination. A child’s score on each of the 25 social skills items was calculated by summing the total number of nominations he or she received from peers. Subsequently, the number was standardized (M = 0, SD = 1) by grade level to control for the different number of students across the grade level units. The standardized scores across the 25 items were then averaged to determine a child’s overall level of social skills. The nomination-based social skills measure had an alpha of .96.

Aggression

The Aggression subscale from the Interpersonal Competence Scale-Teacher report (Cairns et al. 1995; Rodkin et al. 2013) was used to assess children’s aggression. The Aggression subscale consists of three items (“argues,” “gets in a fight,” “gets in trouble at school”), and teachers rated children’s characteristics on a seven-point scale (1 = never; 7 = always). The alpha of the Aggression subscale was .84. Cairns et al. (1995) provided additional reliability and validity evidence of the scale; for example, the Aggression subscale was significantly and moderately related to the direct behavioral observations of negative, hostile behavior as well as peer nominations of hostile behavior.

School Attitudes

Items from the School Sentiment Inventory (Ladd and Price 1987; Valiente et al. 2007) were used to assess children’s school attitudes. The scale included 16 items that tapped into children’s feelings toward school, including enjoyment, loneliness, and dislike (e.g., “Is school enjoyable?” “I feel sad and alone at school. How often does this happen?” “Do you hate school?”). Children rated their feelings on a four-point scale, ranging from 1 (almost never) to 4 (a lot). Negatively worded items were reverse coded to obtain a total score so that a higher score indicated more positive sentiment toward school. The alpha of the School Sentiment scale was .92.

Results

Analysis Overview

Path analysis was used to test the indirect effect of Beliefs about the Importance of Social Skills on School Attitudes via social behaviors (i.e., Aggression, Social Skills). Also, multiple group path analysis was used to examine whether or not gender moderated the associations. First, we screened the data to determine if different statistical assumptions were met. Results indicated that study variables did not meet the assumption of normality of residuals. To address the non-normality issue, we used maximum likelihood estimation with robust standard errors (MLR) in Mplus Version 6.1 (Muthén and Muthén 1998–2010) in path analysis. The significance of indirect effects was tested with Sobel’s test (Sobel 1982); other recommended tests such as a bootstrap method is not available with an MLR estimation.

Descriptive Statistics

Table 1 displays the means, standard deviations, and zero-order correlations for the study variables. The correlations among the study variables were all statistically significant and were in the expected direction. Of particular interest were the correlations between children’s Beliefs about the Importance of Social Skills and their Social Skills, Aggression, and School Attitudes. As shown in Table 1, children who valued social skills more (i.e., higher scores on beliefs about social skills) were reported by peers as more socially skilled, were reported by teachers as less aggressive, and self-reported higher levels of positive school attitudes.

Table 1 Means, standard deviations, and correlations of study variables
Full size table

Indirect Effects of Beliefs about the Importance of Social Skills on School Attitudes Via Social Skills and Aggression

We used path analysis to address our second research goal examining the indirect effects of Beliefs about the Importance of Social Skills on School Attitudes through Social Skills and Aggression. As shown in Fig. 1, a fully saturated model was estimated. Beliefs about the Importance of Social Skills were positively related to social skills which were again positively related to positive School Attitudes. The indirect effect of Beliefs about the Importance of Social Skills on School Attitudes via Social Skills was statistically significant [β = .05, p < .01, 95 % CI (.014, .077)]. That is, children who valued social skills to a greater extent also displayed more socially skillful behaviors, who in turn reported higher levels of positive school attitudes. Beliefs about the Importance of Social Skills were negatively related to Aggression which was again negatively related to positive School Attitudes. The indirect effect of Beliefs about Social Skills on positive School Attitudes via Aggression was also statistically significant [β = .03, p < .05, 95 % CI (.001, .053)]. That is, children who valued social skills to a greater extent displayed less aggressive behaviors, who in turn reported higher levels of positive school attitudes. Overall, the model suggests that Beliefs about the Importance of Social Skills, Social Skills, and Aggression are uniquely associated with School Attitudes, and the variables together explained 16 % of the variance in School Attitudes.

Fig. 1
figure 1

A path model showing that beliefs about the importance of social skills (self-report) are associated with aggression (teacher report) and social skills (peer report), which in turn are related to school attitudes (self-report). **p < .01

Full size image

Gender Effects

We first examined whether or not there are gender differences in the mean levels of the study variables. Independent t tests indicated that girls scored significantly higher than did boys on Beliefs about the Importance of Social Skills, t(329) = −3.44, p < .01, social skills t(332) = −4.28, p < .01, and School Attitudes, t(332) = −2.27, p < .05. In contrast, boys scored significantly higher than did girls on Aggression, t(330) = 4.92, p < .01.

We conducted multiple group path analysis to examine whether or not gender moderated the associations specified in the model. Because the proposed model is just-identified (see Fig. 1), unconstrained models would perfectly fit the data in each group (i.e., boys, girls). By constraining the path coefficients to be equal across boys and girls, however, the model became over-identified and the model fit was able to be evaluated. The constrained model tested a hypothesis that the associations were the same for boys and girls; the model fit the data very well, χ 2(6) = 3.40, p = .76; CFI = 1.00; RMSEA = .00; and SRMR = .04. In other words, the associations specified in Fig. 1 did not differ for boys and girls.

Discussion

The results of the present study demonstrated that elementary school children’s beliefs about the importance of social skills have significant implications for their social behavior and school attitudes. Specifically, children’s beliefs about the importance of social skills were positively associated with social skills and school attitudes and negatively associated with aggression. As hypothesized, beliefs about the importance of social skills had an indirect effect on school attitudes via social behaviors (i.e., social skills, aggression); as children valued social skills more, they displayed higher levels of social skills and lower levels of aggression, and in turn, reported higher levels of positive school attitudes. Gender effects were found in terms of the mean level differences in the study constructs, favoring girls over boys; however, the associations among the study constructs were similar for girls and boys. Notably, the use of multiple sources of informants (i.e., self-report, teacher ratings, peer nominations) increases confidence in the interpretation of the results.

The findings broadly inform the school engagement literature. Thus far, empirical evidence regarding the precursors of elementary children’s school attitudes has been limited to social and behavioral characteristics such as aggression, effortful control, social relatedness (e.g., teacher support, loneliness), and social skills (Gest et al. 2005; Kwon et al. 2012; Valiente et al. 2007). The results of the study showed that children’s social cognition, namely beliefs about the importance of social skills, might be a meaningful precursor of children’s engagement in school. Whereas the focus of the current study was on one aspect of children’s school engagement (i.e., attitude toward school), further research is necessary regarding the degree to which children’s beliefs about the importance of social skills contribute to other aspects of school engagement (e.g., behavioral engagement). Also, it should be noted that the association between beliefs about the importance of social skills and school attitudes is partly attributed to method variance given that both were measured by self-report.

The findings also contribute to the social cognition literature by examining an understudied construct, children’s beliefs about the importance of social skills. Specifically, much research has examined the effect of children’s normative beliefs about aggression on behavioral outcomes (Bellmore et al. 2005; Huesmann and Guerra 1997). This might be intuitive given the detrimental impact of antisocial behaviors on individuals and society. However, there is growing conceptual and empirical evidence that more attention needs to be focused on competence-based behaviors. Conceptually, efforts to simply control aggression might not have as desirable effects without helping children to develop and utilize alternative behaviors conducive to personal and social adjustment (Caprara and Steca 2007). Empirical evidence also showed that socially skillful behaviors significantly contribute to children’s behavioral and academic functioning (Caprara et al. 2000; Chung-Hall and Chen 2010). Findings from this study suggest that children’s beliefs about the importance of social skills potentially serve as a meaningful target for prevention and intervention in one’s effort to promote children’s positive social and academic functioning.

We focused on children’s peer group in our definition of social skills and beliefs about their importance. As children’s social context changes over time, it appears important to conceptualize the constructs in a developmentally sensitive manner (Carlo et al. 2007; Greener and Crick 1999). For example, although socially skillful behaviors are considered to be appropriate in general, the appropriateness of a behavior cannot be adequately determined without fully considering the varying social demands across settings and contexts, including the peer group (Dirks et al. 2007; Sheridan et al. 1999). The peer group serves as an important social context for children to learn and exercise social skills, which in turn facilitates children’s successful interpersonal relationships and school adjustment.

Findings on gender effects were mostly consistent with what has been suggested in the literature. That is, at least in the mid to late elementary years, girls were found to display more advanced social functioning (i.e., beliefs about of the importance of social skills, social skills, and positive school attitudes) than boys. Also consistent with previous findings (Bellmore et al. 2005; Kwon et al. 2012), gender differences were not detected in the associations among the study variables. For example, the magnitude of association between beliefs about the importance of social skills and social behavior (i.e., aggression and social skills) was similar for both boys and girls. This suggests that beliefs about the importance of social skills contribute to aggression and social skills in a similar manner for both boys and girls.

Given that the current study was based on cross-sectional data, neither causal relations nor temporal conclusions among study constructs are implied. Rather, the conclusion might be more appropriately drawn in light of incremental validity of the measures (see Weems and Stickle 2012). Specifically, the association between beliefs about the importance of social skills and school attitudes remained significant after social skills and aggression were included in the path model. This suggests that each of the three variables uniquely accounts for variance in school attitudes and, thus, is useful in understanding the phenomenon.

Limitations and Future Directions

The current study examined one aspect of social cognition (i.e., beliefs about the importance of social skills), and there are other social cognitive factors that might be considerably related to children’s social behaviors and school attitudes. For example, research has found that children with higher levels of agentic goals (goals to gain admiration and respect from others and to have control over others) were perceived by peers to be more aggressive whereas children with higher levels of communal goals (goals for interpersonal closeness and affiliation) were perceived by peers to be more prosocial (Ojanen et al. 2005). Similarly, a recent meta-analytic study has documented negative associations between prosocial goals and aggression and positive associations between antisocial goals and aggression (Samson et al. 2012). Future studies are warranted that examine how children’s beliefs about social skills and social goals work together to affect adjustment outcomes.

Some limitations regarding research methods and designs are also worth noting. First, the instrument used to assess aggression in this study does not represent different forms of aggression documented in the literature. Specifically, as relational aggression is more prevalent among girls than boys (Crick and Grotpeter 1995), future studies can be improved by using a measurement tool that captures different forms of aggression displayed by both boys and girls. Second, given that the participants were predominantly White and drawn from Midwestern rural communities, it was not possible to examine how the relations might vary by culturally and linguistically diverse groups of students. In fact, a previous study found race effects in adolescents’ prosocial values (Beutel and Johnson 2004). Further investigation is necessary regarding the race/ethnicity effect on beliefs about the importance of social skills as well as their relations to outcomes among children in the elementary years. Third, we used quantitative research methods. As such, the results are bounded by the constructs and measures chosen by the researcher. Mixed methods research designs might be a useful avenue to pursue in future research to address this limitation. For instance, children’s narratives might disclose meaningful insights regarding how their beliefs about social skills are related to social behavior and, in turn, school attitudes. Fourth, children’s beliefs about the importance of social skills and school attitudes were measured by self-report and thus could be subject to social desirability in responding. Although we assured children that their answers were private, it is possible that children responded to questions in a manner consistent with what they thought was socially acceptable, rather than their actual feelings.

Implications

The results of the present study have meaningful implications for educational practice in assessment, intervention, and prevention. The findings of the study suggest that children’s beliefs about social skills, along with beliefs about aggression, are important to consider in efforts to foster children’s positive social functioning as well as school engagement. Children’s beliefs about the importance of social skills can be easily incorporated into existing social skills training programs or school-wide positive behavior support programs. For example, school personnel who implement those programs might consider assessing children’s beliefs about the importance of social skills as part of the program (e.g., pretest and posttest) and discuss the values of social skills with children as part of the curriculum. School personnel can also inform parents of the significance of children’s beliefs about the importance of social skills in school-related functioning and encourage their involvement at home. Furthermore, beliefs about the importance of social skills might be particularly useful to identify and help students who dislike school and/or those who are at-risk for school disengagement early in schooling. Finally, the assessment and intervention of children’s beliefs about the importance of social skills might be further informed and targeted by taking into account important social contexts for children (e.g., peer group).