Author: Inglés, Cándido J; Delgado, Beatriz; García-Fernández, José M; Ruiz-Esteban, Cecilia; Díaz-Herrero, Ángela
Date published: July 1, 2010
Journal code: SJPY
Peer interaction is of vital importance for the cognitive, emotional, and social development of children and adolescents. Reciprocal peer relations condition the learning of skills and attitudes towards such interactions (Bukowski, Bredgen, & Vitaro, 2007). Thus, children and adolescents who have poor social relations may have more trouble interacting with other peers, presenting worse psychosocial adjustment and a higher risk of developing psycho-affective disorders in the future (Garaigordobil, 2006).
Ortiz, Aguirrezabala, Apokada, Etxebarria, and López (2002) reported the existence of three important behavioral tendencies in social interaction: other-orientation, defined as prosociability, oriented against the other or aggressiveness, and the tendency to avoid the other, which is typical of withdrawal and social anxiety (Rapee & Sweeney, 2005). Socially anxious children and adolescents avoid and escape from social situations because they produce high levels of anxiety and cause discomfort, whereas aggressive children violate others' rights, insult, threaten, hit and/or criticize their classmates. In contrast, prosocial children tend to have adequate peer relationships, communicating assertively and empathically, and displaying cooperative and helping behaviors in the classroom (Inglés et al., 2008). The prevalence of these social interaction patterns is high in Spanish adolescents. Specifically, the study conducted by Inglés et al. (2008) revealed that 17% of the adolescents were identified as prosocial, and 16% and 12%, respectively, were identified as aggressive adolescents and as adolescents with social anxiety.
Sociometric nomination expresses a student's position or status in the class, according to the opinion of his or her peers. Currently, the classification methodology most broadly accepted by the scientific community is bidimensionality, considering social preference and the impact of social variables from which derive five sociometric types: liked, rejected, neglected, controversial, and average. The prevalence of these sociometric types varies significantly depending on the classification system employed (García-Bacete, 2007; Muñoz, Moreno, & Jiménez, 2008). Recent studies using the probabilistic classification criterion of García-Bacete found a prevalence range of 11.4-11.7% for liked, 12-12.3% for rejected, 12.9-13.7% for neglected, 4-5% for controversial, and 57.9-59.2% for average in groups of students between 10-13 years of age (García-Bacete, 2007; García-Bacete, Sureda, & Monjas, 2008).
Adolescents' interpersonal relations with their classmates have an impact on their degree of acceptance or rejection within the peer group. Most of the empirical studies have focused on the analysis of the cognitive-behavioral patterns that characterize the diverse sociometric types (Bukowski et al., 2007). Peers perceive liked classmates as more sociable, less isolated and aggressive, rejected classmates as more aggressive and slightly isolated, and neglected classmates as less sociable and aggressive and more isolated than their liked counterparts, whereas controversial classmates are perceived like aggressive classmates but as being more sociable than rejected classmates (García-Bacete, 2007; Jiménez, 2003; Muñoz et al., 2008). In addition, rejected and neglected adolescents usually present higher levels of social anxiety than the remaining sociometric types, although neglected students present the highest social inhibition (Inderbitzen, Walters, & Bukowski, 1997). In this sense, a recent study carried out with Spanish preadolescents revealed that the reasons for rejecting a classmate involved behaviors associated with aggressiveness, such as prepotency, dominance, intimidation, and verbal and/or physical aggression, and not so much because of inhibition or social withdrawal (Monjas, Sureda, & García-Bacete, 2008). In this line, diverse predictive studies have partially confirmed these results, finding that low aggressiveness and high sociability are significant predictors of peer liking (e.g., Puckett, Wargo, & Cillessen, 2008) whereas longitudinal studies have identified the sociometric typologies as predictors of future social adjustment, with sociometric popularity or liking in Primary Education appearing as a negative predictor of low aggressiveness and externalizing problems in Secondary Education students (e.g., Mayeux & Cillessen, 2008).
The cognitive-behavioral correlates of the sociometric types vary as a function of gender and, to a lesser degree, of age (Inglés et al., 2008). Thus, girls are usually more prosociability oriented and they develop more empathic strategies, so they are significantly more nominated as liked by their classmates (Pakaslahti, Karjalainen, & Keltikangas-Jarvinen, 2002), although this has not been confirmed in Spanish studies (García-Bacete et al., 2008; Sureda, García-Bacete, & Monjas, 2009). Likewise, the prevalence of social anxiety is usually higher in girls (Inglés et al., 2008; Rapee & Sweeney, 2005), which, along with higher social avoidance and a lower level of aggressiveness (Inglés et al., 2008), predisposes them to be more nominated as neglected by their classmates (García-Bacete et al., 2008). Moreover, boys usually display more aggressive behaviors than their female classmates (Buelga, Musitu, Murgui, & Pons, 2008; Inglés et al., 2008). Therefore, they are rejected to a greater extent by their peers than the girls (García-Bacete et al., 2008).
Regarding age, recent studies have concluded that most of the sociometric types remain stable during the school years. Jiang and Cillessen (2005) concluded, after the meta-analysis of 77 studies, that there was evidence of stability in the accepted and rejected types, with more stability in older children. This indicates that the structure of peer groups becomes stronger as the children grow older and, therefore, peer nomination is also more stable. In this sense, García-Bacete et al. (2008) found that only the rate of controversial and liked girls varied over the courses, increasing and decreasing from Primary Education to 1st grade of Compulsory Secondary Education (CSE), respectively.
Despite the great interest aroused by the sociometric perspective to predict children's and adolescents' future social adjustment, there is an evident lack of works that analyze the relation between the position or social status in the classroom, mainly with regard to the neglected group, and adolescents' self-reported social interaction styles when they interact with their peers. Therefore, this study had the following goals: (a) to analyze the differences in prevalence rates of sociometric types with regard to the patterns of social interaction (prosocial, aggressive, and inhibited), taking into account the variables gender and academic grade in a representative sample of Spanish adolescents; and (b) to examine, by means of logistic regression analysis, the reciprocal predictive capacity of sociometric types and social interaction styles in adolescence.
From previous empirical evidence, the following hypotheses are derived:
1. Prosocial adolescents are more likely to be chosen as liked than aggressive adolescents and adolescents with social anxiety;
2. Aggressive adolescents are more likely to be rejected than prosocial and socially anxious adolescents;
3. Adolescents with social anxiety are more likely to be rejected and neglected than prosocial adolescents; however, they are less likely to be rejected than aggressive adolescents;
4. Prosociability and low aggressiveness will be positive predictors of being liked, whereas aggressiveness will be a negative predictor of being liked; and
5. Peer liking will significantly predict adolescents' social interaction styles. With regard to the sociometric types of rejected and neglected, we expect that both will act as significantly negative predictors of prosociability and positive predictors of aggressiveness and social anxiety. Moreover, we expect that prosociability will act as a negative predictor of being rejected and neglected, whereas aggressiveness and social anxiety will be positive predictors of these sociometric types.
The analysis of the differences of percentages between liked, rejected, and neglected adolescents according to their social interaction style will allow us to: (a) identify the prevalence of sociometrically "typical" students " (likedprosocial and rejected-aggressive); (b) carry out an analysis of the sociometrically "untypical" adolescents (liked-with social anxiety, rejected-prosocial, neglected-prosocial); and (c) to study in more depth the analysis of the neglected students, the sociometric group that has received the least attention. In addition, the use of logistic regression analysis will contribute a novel view of the reciprocal predictive capacity among the sociometric types and the social interaction styles in adolescence, a phenomenon that has yet to be investigated in our country.
Random cluster sampling was carried out (geographical areas of the province of Alicante and the Region of Murcia: centre, north, south, east, and west). A total of 1,419 students participated in this work, from 1st to 4th grade of Compulsory Secondary Education (CSE) (sampling error = .02), of whom 70 (4.93%) were excluded because of errors or omissions in their responses, because they did not obtain their parents' consent to participate in the investigation, or because their mastery of the Spanish language was deficient
The final sample comprised 1.349 students (697 boys and 652 girls): from the 1st grade of CSE (203 boys and 183 girls), 2nd grade of CSE (173 boys and 152 girls), 3rd grade of CSE (172 boys and 146 girls) and 4th grade of CSE (149 boys and 171 girls). Age range was from 12 to 18 years (M = 13.86, SD = 1.38). Using the chi-square test to check for homogeneous distribution of frequencies, we confirmed that there were no statistically significant differences among the eight groups of Gender x Grade (χ^sup 2^ = 4.53, p = .21).
The sociocultural level of the adolescents was recorded via the parents' educational level: 10.82% of the fathers and 11.18% of the mothers had primary studies (School graduate), 66.6% of the fathers and 69.67% of the mothers had middle studies (High school-Pre-university degree), and 16.03% of the fathers and 13.24% of the mothers had university studies (Diploma or Licentiate degree). As some of the participants did not provide information about this variable, we did not have these data for 6.55% of the fathers and for 5.91% of the mothers.
The Teenage Inventory of Social Skills (TISS; Inderbitzen & Foster, 1992). The TISS has 40 items grouped into two scales, Prosocial Behavior, which assesses cooperative, helping, and friendly behaviors (for example, "I offer my classmates help to do their homework") and Antisocial Behavior, which assesses aggressive behaviors, disruptive reactions, and attention seeking (for example, "I hit other kids when they make me mad"). The items are rated on a 6-point Likert scale ranging from 1 (it doesn't describe me at all) to 6 (it describes me completely).
The psychometric properties of the TISS in North American adolescent population were satisfactory (Inderbitzen & Foster, 1992) and similar to those found in a sample of Spanish adolescents (Inglés, Hidalgo, Méndez, & Inderbitzen, 2003). In this study, the internal consistency coefficients (Cronbach's alpha) were satisfactory for both scales: Prosocial Behavior (.89) and Antisocial Behavior (.83).
The Social Phobia and Anxiety Inventory (SPAI; Turner, Beidel, Dancu, & Stanley, 1989). The SPAI is the most frequently employed questionnaire to assess social anxiety in adolescence. It has 45 items that measure social phobia and agoraphobia by means of two subscales. The Social Phobia subscale has 32 items (for example, "I feel nervous when I am in social situations where there is a big group of people"), rated on a 7-point scale, ranging from 1 (never) to 7 (always). The SPAI presents excellent reliability and validity in adolescent Spanish population (García-López, Piqueras, Díaz-Castela, & Inglés, 2008). For the goals of this study, we only used the Social Phobia subscale, because it has been identified as the most precise and specific measure to assess responses of social anxiety in Spanish adolescents (Olivares et al., 2002). In this study, the Social Phobia subscale presented a satisfactory internal consistency index (.95).
Sociometric Peer Nomination Test. The sociometric test allows one to determine the level of acceptance or rejection of the members of a group, to discover the relations among the individuals, and to reveal the structure of the group in order to identify the liked, rejected, and neglected members. We used the probabilistic nomination procedure with three inter-gender choices, considered the most adequate and precise of the sociometric nomination tests (García-Bacete, 2007). The following items were included in the questionnaire: (1) Write the name of three classmates whom you like the most; and (2) Write the name of three classmates whom you like the least. This study only focused on the analysis of the liked, rejected, and neglected subjects because these make up the largest number of students (García-Bacete) and, in turn, they represent the best (liked) and worst social adjustment (rejected and neglected) in the academic setting.
Once we had obtained the approval of the headmasters and the psycho pedagogues of the centers and the written informed consent of the participants' parents to take part in this investigation, we administered the tests. The questionnaires were completed collectively, voluntarily, and anonymously in the classrooms. The investigators read the instructions of each test out loud, underlining the importance of responding to all the items. The mean duration of each instrument was 15 minutes.
The students' sociometric identification was carried out by means of the Program Socio (González, 1990) by which one obtains the lower and upper limits of the positive nominations received (LL [Pn] and UL [Pn]) and of the negative nominations received (LL [Nn] and UL [Nn]) for a group or class of students, by calculating the binomial probability to find the value of the t test associated with a certain skewness and a level of probability of less than .05 (see Salvosa tables in Bastin, 1966). Identification was made according to the following criteria: Liked = Pn ≥ UL (Pn) and Nn < M (Nn); Rejected = Nn ≥ UL (Nn) and Pn < M (Pn); and Neglected = Pn ≤ 1 and Nn < M (Nn). We identified 201 (14.9%) liked students, 162 (12%) rejected students, and 69 (5.1%) neglected students.
To identify students with social anxiety, we used the cut-off score equal to or higher than 100 (rate of false positives = 1.32%) on the Social Phobia subscale of the SPAI (Olivares et al., 2002), whereas to identify prosocial and aggressive students, we used the cut-off score of the mean plus one standard deviation on the Antisocial Behavior and Prosocial Behavior scales of the TISS. We chose these cut-off points because the SPAI assesses a clinical construct and has a reliable cut-off point for the identification of subjects with high social anxiety, whereas the TISS assesses non-clinical constructs. Thus, we identified 163 students with social anxiety (12.1%), 232 aggressive students (17.2%), and 229 prosocial students (17%).
To check for statistically significant differences among the proportions of sociometric types and social interaction styles, we applied Cochran's nonparametric Q test, which allows the comparison of K related proportions, for the total sample, by gender and grade. In the cases in which the Q test detected statistically significant differences, we applied the Z test for differences in proportions to find between which sociometric types or social interaction styles such differences occurred. Also, in order to eliminate possible effects of sample size on the statistical analyses, we calculated the effect size or mean standard difference (d index) which determines the magnitude of the differences found by the Z test. The interpretation of the effect size is simple: values lower than or equal to .20 indicate a very small effect size; values between .20 and .49, a small one; between .50 and .79, a moderate effect size; and values higher than .80, a large effect size (Cohen, 1988).
Lastly, predictor equations of the sociometric types and the social interaction styles were formulated with forward stepwise logistic regression analysis based on Wald's statistic, as the variables assessed in the study are categorical and do not meet the assumptions of the general linear model. The fit of the models was carried out by means of Nagelkerke's R^sup 2^.
Prevalence of Sociometric Types among Students with Social Anxiety, Aggressive, and Prosocial Students
Table 1 presents the relative frequencies and the proportions corresponding to the interaction between sociometric types (students identified as liked, rejected, and neglected) and social interaction styles (students with social anxiety, aggressive, and prosocial students). The proportions ranged between .2% (prosocial-neglected students) and 3.6% (prosocial-liked students).
The Q statistic revealed statistically significant differences in the prevalences of social interaction styles in liked, rejected, and neglected students. Specifically, the Z test detected that the proportion of liked students with social anxiety was significantly lower than the proportion of liked-aggressive and liked-prosocial students. In addition, the proportion of rejected-aggressive students was significantly higher than the proportion of rejected students with social anxiety and rejected-prosocial students. Lastly, the rate of neglected students with social anxiety was statistically higher than the rate of neglected-prosocial students, whereas the rate of neglected-aggressive students was significantly higher than that of neglected-prosocial students. The magnitude of the differences found was, in all cases, lower than .15, a value that indicates a very small effect size (Cohen, 1988).
With regard to gender, the Q test detected prevalence differences in the social interaction styles of the liked, rejected, and neglected boys and of the liked girls (see Table 2). The results indicate that the proportion of likedaggressive boys was significantly higher than the proportion of liked boys with social anxiety and of liked-prosocial boys. This pattern of results was repeated in the sociometric types of rejected and neglected boys. Regarding the girls, only two statistically significant comparisons were found. Specifically, the rate of liked-prosocial girls was higher than that of liked-aggressive girls and of liked girls with social anxiety. In all cases, the effect sizes of the differences found were low in magnitude (d < .21).
As seen in Table 3, the Q test only detected statistically significant differences in the rejected adolescents of 3rd grade of CSE and in the liked and neglected adolescents of 4th grade of CSE. Specifically, in 3rd grade of CSE, the proportion of rejected-aggressive students was significantly higher than that of rejected classmates with social anxiety (d = -.19) and of rejected-prosocial classmates (d = .14). In 4th grade of CSE, the results indicated that the proportion of liked adolescents with social anxiety was significantly lower than that of liked-prosocial (d = -.26) and liked-aggressive (d = -.14) adolescents. Lastly, the rate of neglected-prosocial adolescents was significantly lower than that of neglectedaggressive (d = .18) and neglected adolescents with social anxiety (d = .13).
Prevalence of Social Interaction Styles among Liked, Rejected, and Neglected Students
The Q tests detected statistically significant differences in the prevalences of sociometric types in aggressive adolescents (Q = 7.23, p = .03) and prosocial adolescents (Q = 45.65, p = .00). The results indicate that the proportion of aggressive-liked students (Z = 2.69, p = .00, d = .10) and of aggressive-rejected students (Z = 2.23, p = .01, d = .09) was significantly higher than that of aggressive-neglected students. However, the percentage of prosocial-liked students was significantly higher than that of prosocialrejected students (Z = 3.87, p = .00, d = .50) and of prosocialneglected students (Z = 6.48, p = .00, d = .25), whereas the proportion of prosocial-rejected students was higher than that of prosocial-neglected students (Z = 3.25, p = .00, d = .13). No statistically significant differences were found in the remaining comparisons among sociometric types in adolescents with social anxiety and aggressive adolescents.
As a function of gender, the Q statistics detected statistically significant differences in the prevalence rates of sociometric types in prosocial boys (Q = 8.40, p = .02) and boys with social anxiety (Q = 13, p = .00), as well as in prosocial girls (Q = 39, p = .00). In addition, the results indicated that the rate of prosocial-neglected boys was significantly lower than that of prosocial-liked boys (Z = 3.03, p = .00, d = .16) and prosocial-rejected boys (Z = 2.57, p = .00, d = .13). Likewise, the proportion of rejected boys with social anxiety was significantly higher than that of liked boys with social anxiety (Z = -2.21, p = .01, d = -.12) and neglected boys with social anxiety (Z = 3.37, p = .00, d = .18). Regarding the girls, the prevalence of prosocialliked girls was significantly higher than that of prosocialneglected girls (Z = 5.62, p = .00, d = .31) and of prosocialrejected girls (Z= 3.91, p = .00, d = .22), and the rate of prosocial-rejected girls was significantly higher than that of prosocial-neglected girls (Z = 2.20, p = .01, d = .12).
With regard to academic grade, the results revealed that most of the prevalences of social interaction styles and sociometric nomination were invariant across all the grades analyzed. Nevertheless, the Q tests found statistically significant differences of a low magnitude (d < .33) in the prevalences of prosocial students from 1st, 3rd, and 4th grade of CSE. Specifically, the prevalence of prosocial-liked students was significantly higher than that of: (a) prosocialneglected students in 1st grade of CSE (Z = 2.98, p = .00, d = .22); (b) prosocial-neglected students (Z = 3.91, p = .00, d = .31); (c) prosocial-rejected students (Z = 2.24, p = .01, d = .18) in 3rd grade of CSE; and d) prosocial-neglected students (Z = 2.30, p = .01, d = .33) and prosocial-rejected students (Z = 2.57, p = .00, d = .23) in 4th grade of CSE. Likewise, the prevalence of prosocial-rejected students was significantly higher than that of prosocial-neglected students in 1st grade of CSE (Z = 2.25, p = .01, d = .16), 3rd grade of CSE (Z = 2.28, p = .01, d = .18), and 4th grade of CSE (Z = 4.17, p = .00, d = .16).
Prediction of Social Interaction Styles
The criterion variables corresponding to the adolescents' social interaction styles were dichotomized, categorizing them by the above-mentioned cut-off points, in prosocial (n = 69) and not prosocial (n = 144), aggressive (n = 93), and not aggressive (n = 120), and with social anxiety (n = 51), and without social anxiety (n = 120). Although the logistic regression analyses did not allow us to create a model to predict social anxiety, we obtained two regression models to predict prosociability and aggressiveness. Thus, the model created to predict prosociability allowed the correct estimation of 83% of the cases χ^sup 2^ = 21.25, p = .00, with the variables being neglected, liked, and rejected forming part of the equation (see Table 4). Moreover, the model created to predict aggressiveness allowed the correct estimation of 82.8% of the cases χ^sup 2^ = 4.83, p = .03) with the variable being neglected forming part of the equation. In this sense, the models obtained fit values (R^sup 2^ Nagelkerke) of .03 and .01, respectively.
The odds ratio (OR) obtained in the model that predicts prosociability indicated that the liked students are 48% more likely to be prosocial, whereas the neglected and rejected adolescents are 79% and 41% less likely to be prosocial, respectively. With regard to the model to predict aggressiveness, the OR indicated that the neglected adolescents were 90% more likely to be aggressive.
Prediction of Sociometric Types
In this case, the criterion variables were categorized by means of the criteria proposed in the Program Socio: liked (n = 105) and not liked (n = 108), rejected (n = 75) and not rejected (n = 138), and neglected (n = 33) and not neglected (n = 180). In addition, the predictor variables were dichotomized as being or not being prosocial, being or not being aggressive, and having or not having high social anxiety. Thus, we could create three logistic models to predict the variables of being chosen by the classmates as liked, rejected, and neglected (see Table 4). The model created to predict being liked (or popularity) allowed the correct estimation of 85.1% of the cases (χ^sup 2^ = 7.41, p = .01), and the variable prosociability was entered in the equation. With the variable predictor prosociability, the model created to predict being rejected also allowed the correct estimation of 88% of the cases (χ^sup 2^ = 4.93, p = .03). Lastly, the model created to predict being neglected allowed the correct estimation of 94.9% of the cases (χ^sup 2^ = 15.16, p = .00), and the variables prosociability and aggressiveness were entered in the equation. With regard to the fit value of the models, we obtained a Nagelkerke R^sup 2^ of .01 for the model of liked and rejected and of .03 for that of neglected.
The ORs obtained in the models indicated that: (a) prosocial adolescents are 67% more likely to be chosen as liked, 42% less likely to be rejected, and 78% less likely to be neglected by their classmates; (b) aggressive adolescents are 83% more likely to be chosen as neglected by their classmates.
The purpose of the present study was to examine the relation between social interaction styles (aggressiveness, prosociability, and social anxiety) and sociometric types (liked, rejected, and neglected) in a sample of Spanish adolescents.
The results revealed that prosocial students were proportionately chosen more by their classmates as liked than students with social anxiety, partially confirming the first hypothesis, because, surprisingly, no statistically significant differences between the rate of liked-prosocial adolescents (3.6%) and liked-aggressive adolescents (2.9%) were obtained. This result could be explained in two ways. Firstly, because of the discrepancy among the measures of sociability and aggressiveness employed in the present study (self-report measures) and those used in previous studies (peer nomination). Secondly, the age range of the students in this work (CSE) was different from that analyzed in previous studies (Primary Education). In this sense, it has been reported that aggressive behaviors are more direct in infancy and preadolescence, and more indirect in adolescence (Heilbron & Prinstein, 2008), which could affect peers' social preference. Bearing in mind both these aspects, future work should include different sources of assessment of social behavior (self-reports, peers, teachers, and parents) to analyze inter-source concordance. Moreover, future studies should be more precise in the measure of aggressive behavior, assessing the diverse forms of expression (physical, relational, and indirect) and the reasons that cause it (proactive and reactive).
The rate of prosocial-liked students was significantly higher than that of prosocial-rejected and prosocialneglected students, in accordance with previous results, where the prosocial students were more accepted and less rejected by their classmates (García-Bacete, 2007; Muñoz et al., 2008), and the most important reasons for liking a classmate were positive and prosocial characteristics such as fellowship, being nice, fun, and a good friend (Monjas et al., 2008).
As established in the second hypothesis, aggressive students were rejected the most by their classmates. These results support those obtained in previous investigations in which aggressive behaviors were significant factors of lower social acceptance and higher peer rejection (García- Bacete, 2007; Monjas et al., 2008). Once again, these results underline the importance of paying special attention to this group of students because they have worse levels of academic and family self-esteem and they display more internalizing symptoms, becoming a high risk group for academic maladjustment (Estévez, Herrero, Martínez, & Musitu, 2006).
The students identified as having social anxiety were the chosen as the least liked among their classmates, and moreover, they were also rejected and neglected more than the prosocial students. This result could be due to these adolescents' characteristic pattern of social interaction because, in their attempts to avoid social situations, they are more likely to be less visible, less valued, and rejected more by their peers (Inderbitzen et al., 1997). In this sense, in accordance with the proposal of Bukowski et al. (2007), it seems that the characteristics that lead to an adolescent being rejected not only depend on aggressiveness, but instead on other predictor variables such as social anxiety, negative emotions, lack of confidence, lack of tolerance, isolation, and the lack of prosocial and cooperative behaviors in classroom may also intervene. Nevertheless, students with social anxiety were rejected less than aggressive students, thus confirming that aggressive and disruptive behaviors are the most important reasons for rejecting a classmate, whereas social withdrawal is a less relevant explanation of peer rejection (Monjas et al., 2008).
Regarding the prevalence differences by gender, the analyses revealed that aggressive boys were the most highly represented in the three sociometric groups, with the boys with social anxiety being the least liked and the prosocial boys the least rejected and neglected. Therefore, the results of this study provide new findings concerning previous studies because, although the studies performed with both sexes found that rejected adolescents were perceived as aggressive by their peers, and liked adolescents were generally perceived as prosocial (García- Bacete, 2007, Muñoz et al., 2008), the results of the present study indicate that aggressive boys are equally nominated as rejected, liked, and neglected. This result could be due to the fact that aggressive students do not only have critics but, in most cases, they also have a reduced support group that reinforces their aggressive and deviant behaviors, thus maintaining some degree of popularity. With regard to the girls, the fact that the percentage of liked and prosocial girls is higher than that of liked girls with social anxiety and liked-aggressive girls reinforces the lines of previous findings that indicate a higher tendency in female adolescents towards prosociability (Inglés et al., 2008), empathy, and assertion, highly valued reasons for peer acceptance (Monjas et al., 2008).
Regarding the differences in academic grade, this study reveals some stability in the percentages of social interaction styles and sociometric types, finding only statistically significant differences, similarly to the findings for the total sample, for the adolescents of the second cycle of CSE. Thus, the rate of rejected and aggressive students in 3rd grade of CSE was significantly higher than that of rejected students with prosocial characteristics and with social anxiety. The prosocial adolescents were the most popular in 1st, 3rd, and 4th grades of CSE and the least neglected in 4th grade of CSE. This study indicates that prosocial adolescents are more liked and less neglected. In contrast, students with social anxiety from 4th grade of CSE are significantly less frequently chosen as liked than their prosocial and aggressive classmates. This result reveals the negative impact of behavioral inhibition on group acceptance in adolescence, a stage with higher demands of interaction and social openness (Bukowski et al., 2007).
Lastly, by means of logistic regression analyses, we created five predictive models with a high percentage of correctly classified cases (82.8-94.9%), by which we confirmed the predictive and reciprocal relation of social interaction styles and adolescents' reputation with their classmates. As stated in the fourth hypothesis, adolescents' prosocial behaviors were a protector factor against being rejected and neglected, increasing the likelihood of being liked by their peers. In contrast, adolescents' aggressive behaviors act as a positive predictor of being neglected by their peers. Moreover, peer acceptance acted as a positive predictor of prosocial behavior. In contrast, being rejected and neglected by the peer group decreased the likelihood of behaving cooperatively and prosocially. Likewise, being neglected by the peer group was a positive predictor of aggressive behaviors in the classroom. These findings confirm the predictive power of cooperative behaviors in the classroom for acceptance and preference (Puckett et al., 2008), and, more important, they identify relations not previously analyzed, underlining the predictive role of student's status in the classroom for the display of socially adaptive and maladaptive behaviors with the classmates. Therefore, teachers and psychopedagogues could use sociometric types as a tool to identify possible deficits in interpersonal skills.
The results of this investigation are important for two reasons. Firstly, the study confirms the existence of a significant relation between social interaction styles and the adolescent's social position in the classroom, especially, between prosociability and popularity and aggressiveness and rejection. However, upon examining in more depth the connection between the latter two variables, a pure relation between sociometric type and social interaction style is not found. Thus, there is a high percentage of aggressive students who are chosen as neglected by their classmates. Secondly, the results of this work reveal the importance of the variable gender when relating and being liked, rejected, or neglected by the peer group in adolescence. In this sense, and along the lines of previous research (García-Bacete, et al., 2008; Pakaslahti et al., 2002), aggressiveness in boys and isolation and prosociability in girls considerably establish their social status among their classmates.
Despite the fact that prosociability is more frequent among adolescents (Inglés et al., 2008), this study reveals the existence of a high percentage of students who are at risk in the classroom (rejected and neglected), who need support and professional intervention to improve their psychosocial adjustment. Likewise, efforts should be made to apply and assess programs aimed at promoting behaviors that could substantially improve adolescents' sociometric status.
The results of the study should be interpreted taking into account some limitations, which could be remedied in future research. Firstly, the diverse groups of neglected and rejected students should be analyzed (Estévez et al., 2006; Inderbitzen et al., 1997), as well as the average and controversial sociometric categories, because the omission of these analyses could lead to incomplete results when classifying the students in a certain sociometric type. In addition, as indicated above, future research should use tests that differentiate the diverse types of aggressiveness because they can affect sociometric status differentially, through adolescents' gender and age. Likewise, longitudinal studies should be carried out during the entire pupil age to establish causal relations among variables, using the structural equation modelling. Social functioning in the classroom should also be analyzed, taking into account additional sources of information (for example, teachers and parents), as well as teachers' beliefs and attitudes about their students' social behavior, and about the cultural and contextual characteristics of the assessment environment, because these variables can act as important mediators in students' sociometric choice (Bukowski et al., 2007).
Bastin, G. (1966).Los tests sociométricos [Sociometric tests]. Buenos Aires: Kapelusz.
Buelga, S., Musitu, G., Murgui, S., & Pons, J. (2008). Reputation, loneliness, satisfaction with life and aggressive behavior in adolescents. The Spanish Journal of Psychology, 11, 192-200.
Bukowski, W. M., Brendgen, M., & Vitaro, F. (2007). Peers and socialization: Effects on externalizing and internalizing problems. In J. E. Grusec, & P. D. Hasting (Eds.), Handbook of socialization: Theory and research (pp. 355-381). New York: Guilford Press.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Estévez, E., Herrero, J., Martínez, B., & Musitu, G. (2006). Aggressive and nonaggressive rejected students: An analysis of their differences. Psychology in the Schools, 43, 387-400.
Garaigordobil, M. (2006). Psychopathological symptoms, social skills and personality traits: A study with adolescents. The Spanish Journal of Psychology, 9, 182-192.
García-Bacete, F. J. (2007). La identificación de los alumnos rechazados, preferidos, ignorados y controvertidos en el aula [The identification of rejected, liked, neglected, and controversional students in the classroom]. Revista de Psicología General y Aplicada, 60, 25-46.
García-Bacete, F. J., Sureda, I., & Monjas, I. (2008). Distribución sociométrica en las aulas de chicos y chicas a lo largo de la escolaridad [Sociometric distribution of boys and girls during all the schooling period]. Revista de Psicología Social, 23, 63-74.
García-López, L. J., Piqueras, J. A., Díaz-Castela, M. M., & Inglés, C. J. (2008). Trastorno de ansiedad social en la infancia y adolescencia: Estado actual, avances recientes y líneas futuras [Social anxiety disorder in childhood and adolescence: Current state, recent advances, and future lines]. Behavioral Psychology/Psicología Conductual, 16, 501-532.
González, J. (1990). Sociometría por ordenador [Sociometry by computer]. Valencia: Consellería de Cultura, Educació i Ciencia.
Inderbitzen, H., & Foster, S. L. (1992). The Teenage Inventory of Social Skills: Development, reliability, and validity. Psychological Assessment, 4, 451-459.
Inderbitzen, H., Walters, K., & Bukowski, A. (1997). The role of social anxiety in adolescent peer relations: Differences among sociometric status groups and rejected subgroups. Journal of Clinical Child Psychology, 26, 338-348.
Inglés, C. J., Martínez-Monteagudo, M. C., Delgado, B., Torregrosa, M. S., Redondo, J., Benavides, G., et al. (2008). Prevalencia de la conducta agresiva, conducta prosocial y ansiedad social en una muestra de adolescentes españoles: Un estudio comparativo [Prevalence of aggressive behavior, prosocial behavior, and social anxiety in a sample of Spanish adolescents: A comparative study]. Infancia y Aprendizaje, 31, 449-461.
Inglés, C. J., Hidalgo, M. D., Méndez, F. X., & Inderbitzen, H. M. (2003). The Teenage Inventory of Social Skills: Reliability and validity of the Spanish translation. Journal of Adolescence, 26, 505-510.
Heilbron, T., & Prinstein, M. J. (2008). A review and reconceptualization of social aggression: Adaptive and maladaptive correlates. Clinical Child and Family Psychology Review, 11, 176-217.
Jiang, X. L., & Cillessen, A. H. (2005). Stability of continuous measures of sociometric status: A meta-analysis. Developmental Review, 25, 1-25.
Jiménez, M. (2003). Características emocionales y comportamentales de los grupos sociométricos desde una perspectiva multiple. [A multiperspective of emotional and behavioral correlates of sociometric groups]. Psicología Conductual, 11, 41-60.
Mayeux, L., & Cillessen, A. H. (2008). It's not just being popular, it's knowing it, too: The role of self-perception of status in the associations between peer status and aggression. Social Development, 17, 871-888.
Monjas, M. I., Sureda, I., & García-Bacete, F. J. (2008). ¿Por qué los niños y las niñas se aceptan y se rechazan? [Why do children accept and reject each other?]. Cultura y Educación, 20, 479-492.
Muñoz, V., Moreno, M. C., & Jiménez, I. (2008). Las tipologías de estatus sociométrico durante la adolescencia: Contraste de distintas técnicas y fórmulas para su cálculo. [Sociometric typologies during adolescence: Contrasting different scoring techniques and formulas]. Psicothema, 20, 665-671.
Olivares, J., García-López, L. J., Hidalgo, M. D., La Greca, A. M., Turner, S. M., & Beidel, D. C. (2002). A pilot study on normative data for two social anxiety measures: The Social Phobia and Anxiety Inventory and the Social Anxiety Scale for Adolescents. International Journal of Clinical and Health Psychology, 2, 467-476.
Ortiz, M. J., Aguirrezabala, E., Apokada, P., Etxebarria, I., & López, F. (2002). Características emocionales, funcionamiento social y satisfacción en escolares [Emotional characteristics, social functioning and social satisfaction in school age children]. Infancia y Aprendizaje, 25, 195-208.
Pakaslahti, L., Karjalainen, A., & Keltikangas-Jarvinen, L. (2002). Relationships between adolescent prosocial problemsolving strategies, prosocial behaviour, and social acceptance. International Journal of Behavioral Development, 26, 137-144.
Puckett, M. B., Wargo, J., & Cillessen, A. H. (2008). Moderators of the association between relational aggression and perceived popularity. Aggressive Behavior, 34, 563-576.
Rapee, R., & Sweeney, L. (2005). Social phobia in children and adolescents: Nature and assessment. In W. R. Crozier, & L. E. Alden (Eds.), The essential handbook of social anxiety for clinicians. West Sussex, UK: John Wiley & Sons.
Sureda, I., García-Bacete, F. J., & Monjas, M. I. (2009). Razones de los niños y niñas de diez y once años para preferir o rechazar a sus iguales [Reasons why ten and eleven years old children prefer or reject their peers]. Revista Latinoamericana de Psicología, 41, 305-321
Turner, S. M., Beidel, D., Dancu, C. V., & Stanley, M. A. (1989). An empirically derived inventory to measure social fears and anxiety: The Social Phobia and Anxiety Inventory. Psychological Assessment, 1, 35-40.
Received May 10, 2009
Revision received December 20, 2009
Accepted January 19, 2010
Cándido J. Inglés1, Beatriz Delgado1, José M. García-Fernández2, Cecilia Ruiz-Esteban3, and Ángela Díaz-Herrero3
1 Universidad Miguel Hernández de Elche (Spain)
2 Universidad de Alicante (Spain)
3 Universidad de Murcia (Spain)
This work was carried out via the Research Project SEJ 2004-07311/EDUC of the Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica del Ministerio de Educación y Ciencia awarded to the first author.
Correspondence concerning this article should be addressed to Cándido J. Inglés. Área de Psicología Evolutiva y de la Educación. Dpto. de Psicología. Universidad Miguel Hernández. Avda. de la Universidad, s/n. 03202 Elche. Alicante. (Spain). E-mail: firstname.lastname@example.org