Peer relationships during late childhood in internationally adopted and institutionalized children

Depósito de Investigación de la Universidad de Sevilla https://idus.us.es/ Esta es la versión aceptada del artículo publicado en: This is an accepted manuscript of a paper published in: Social development, Volume 30, Issue 1, February 2021, Pages 171–186 DOI: https://doi.org/10.1111/sode.12467 Copyright © 2020 John Wiley & Sons, Ltd. El acceso a la versión publicada del artículo puede requerir la suscripción de la revista. Access to the published version may require subscription. This is the peer reviewed version of the following article: Cáceres, I., Román, M., Moreno, C., Bukowski, W. M., & Palacios, J. (2021). Peer relationships during late childhood in internationally adopted and institutionalized children. Social Development, 30(1), 171–186. which has been published in final form at https://doi.org/10.1111/sode.12467. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. This article may not be enhanced, enriched or otherwise transformed into a derivative work, without express permission from Wiley or by statutory rights under applicable legislation. Copyright notices must not be removed, obscured or modified. The article must be linked to Wiley’s version of record on Wiley Online Library and any embedding, framing or otherwise making available the article or pages thereof by third parties from platforms, services and websites other than Wiley Online Library must be prohibited. PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 1 1 2 Peer Relationships During Late Childhood in Internationally Adopted and 3 Institutionalised Children 4 Isabel Cáceres1, Maite Román1, Carmen Moreno1, William Bukowski2 and Jesús 5 Palacios1 6 1University of Seville (Spain) 7 2Concordia University (Canada) 8 9 Conflict of Interest Statement: The authors declare no conflict of interest. 10 Data Availability Statement: The data that support the findings of this study are 11 available from the corresponding author upon reasonable request. 12 Acknowledgement: This work was supported by the Spanish Ministry of Economy, 13 Industry and Competitiveness and the European Fund FEDER, through two research 14 projects coded PSI2010-19287 and PSI2015-67757-R. Isabel Cáceres’ work was 15 supported by a predoctoral grant funded by the Ministry of Economy, Industry and 16 Competitiveness. The authors are grateful to all the children, parents, caregivers and 17 schools that took part in the study. 18 19 Corresponding author: Isabel Cáceres. Department of Developmental and Educational 20 Psychology, University of Seville. Calle Camilo J. Cela s.n. 41018 Seville, Spain. 21 [email protected]s 22 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 2 Abstract 23 The aim of this study is to analyse the interpersonal relationships in the school context 24 of children living in different care settings (adoptive families, residential care centres, 25 birth families). Participants were 76 children between eight and fourteen years of age 26 (M = 10.78, SD = 1.38), belonging to one of three groups: international adoptees, 27 children living in residential care in Spanish institutions and a comparison sample of 28 Spanish children living with their birth families with no connection with child 29 protection. Sociometric information was collected in the classroom of each child during 30 school hours. Internationally adopted children from Russia showed considerable 31 difficulties in their relationships with peers, they were more likely to be rejected and 32 their peers described them as less prosocial and somewhat more aggressive. With a 33 better sociometric position than the adoptees, children in residential care were rated by 34 their peers as more aggressive and less prosocial than the normative population. 35 Coordination and integrated work between the family, protection centres, schools and 36 other social services should be a strategic priority in the promotion of healthy social 37 development in these groups of children. 38 39 Keywords: international adoption; residential care; peer relations; social 40 relationships 41 42 43 44 45 46 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 3 Most children are born into families that care for them as much as they can. In 47 this family context, children live experiences and create bonds that will influence how 48 they interpret the world and their relationships with others. Children exposed to security 49 and sensitive care in early childhood tend to be more socially competent in the school 50 years (Waters & Sroufe, 1983). On the contrary, when early experiences are based on 51 distrust, neglect or maltreatment, children will have more difficulty creating positive 52 and effective friendships later on. Evidence supporting this model in biological families 53 is abundant (e.g. Groh et al., 2014; Pallini, Schneider, Baiocco, Madigan, & Atkinson, 54 2014) and recent studies have started to address the relations between early adversity 55 and social relationships in adoptive families (e.g. Barone, Lionetti, & Green, 2017). 56 This article extends this analysis also to children living in institutional care. While 57 adopted and institutionalised children have early adversity in common, their life 58 trajectories are very different thereafter. Our goal here is to study the impact of these 59 trajectories (adoption vs. institutional care) on peer relationships in comparison with a 60 normative group. 61 Peer Relationships of Internationally Adopted Children 62 The meta-analysis on peer relationships carried out by DeLuca, Claxton and van 63 Dulmen (2018) included adoptive samples with a wide range of ages and of pre64 adoptive experiences. The results indicated that 82% of adoptees have no significant 65 difficulties in their peer relationships. Although they are less likely to have good 66 friends, the quality of their friendship is similar to that of normative samples. 67 In their study with school-age children, Stams, Juffer, Rispens and Hoksbergen 68 (2000) explored the sociometric status of children adopted from Sri Lanka, South Korea 69 and Colombia in their first weeks of life with little or no institutional exposure. Similar 70 to what happens in normative samples (Cillessen & Bukowski, 2018), 26% were 71 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 4 classified as popular, 52% had an average status, 10% were ignored, 7% were 72 controversial and 5% were rejected. 73 Research has documented the negative impact of early institutional deprivation 74 on later social integration (Gunnar, Van Dulmen, & The International Adoption Project 75 Team, 2007; Sonuga-Barke, Schlotz, & Kreppner, 2010). Moreover, a longer exposure 76 to institutional rearing is related to more peer difficulties (Kadlec & Cermak, 2002; 77 Marcovitch et al., 1997). In particular, studies on the social development of 78 international adoptees from Eastern Europe exposed to institutionalisation tend to report 79 problems in social relationships. According to parents’ and, especially, teachers’ 80 reports, they present greater difficulties in their social functioning than their non81 adopted peers (Caprin, Benedan, Ballarin, & Gallace, 2017; Glennen & Bright, 2005; 82 Petranovich, Walz, Staat, Chiu, & Wade, 2016). These studies indicate that 20% to 30% 83 of post-institutionalised international adoptees experience significant problems with 84 friends or peers (Merz & McCall, 2010; Palacios, Moreno, & Román, 2013). 85 Research on post-institutional children has indicated that conclusions based on 86 self-reports of interpersonal relationships tend to be similar to those of normative 87 samples (Barcons-Castel, Fornieles-Deu, & Costas-Moragas, 2011; Goodman & Kim, 88 2000; Hawk & McCall, 2014). Also, according to Julian and McCall (2016), due to 89 increased social and behavioural demands, the interpersonal problems of post90 institutional children are more evident when assessed in adolescence than in childhood. 91 In particular, research has documented more problems in early adolescence than in later 92 years (Howard, Smith, & Ryan, 2004). 93 Peer Relationships of Children in Residential Care 94 Social development of children in residential care (RC) has been less explored, 95 in part because in many western countries children’s institutionalisation is exceptional 96 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 5 (e.g., in England, Department for Education, 2018). However, in other countries, 97 institutionalisation affects significant percentages of children in care. This is the case of 98 Spain, with 48% of them placed in institutional settings (Ministerio de Sanidad, 99 Consumo y Bienestar Social, 2018), a percentage that is even higher in other countries 100 such as Portugal, Latin-American and Middle-East countries. 101 The studies on children in RC generally show significant difficulties in their 102 social development, with caregivers and teachers reporting more social and relational 103 difficulties than in normative peers (Garcia Quiroga, Hamilton-Giachritsis, & Ibañez, 104 2017; Simsek, Erol, Öztop, & Münir, 2007; Zhang, Cecil, Barker, Mori, & Lau, 2018). 105 Martín, Muñoz De Bustillo, Rodríguez and Pérez (2008) found that, compared with 106 normative samples, children from Spanish institutions were more likely to be rejected 107 and less likely to be chosen by their peers to perform academic tasks. Also, their 108 classmates described them more negatively, as more aggressive and less able to resolve 109 conflicts. Other studies have shown that institutionalised children tend to be more 110 indiscriminate and less selective in their friendship (Roy, Rutter, & Pickles, 2004; 111 Vorria, Rutter, Pickles, Wolkind, & Hobsbaum, 1998) and, despite that, they have fewer 112 reciprocal friends at school than children living in family contexts (Argumendo & 113 Albornoz, 2006; Martín, Muñoz de Bustillo, & Pérez, 2011). Moreover, their social 114 support networks are weaker (e.g., fewer support persons, less help provided) than those 115 of their peers (Bravo & Fernández del Valle, 2003; Makanui, Jackson, & Gusler, 2019; 116 Singstad, Wallander, Lydersen, Wichstrøm, & Kayed, 2019). 117 The comparison of adopted children with those reared in institutions is 118 especially relevant for our purposes. In these comparisons, children in family contexts 119 have better cognitive development, school performance, mental health and fewer 120 behavioural problems than their siblings or peers growing in residential care 121 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 6 (Christoffersen, 2012). Studying the same samples reported here, Palacios et al. (2013) 122 observed that at ages four to eight years children in residential care had more social 123 difficulties and less social skills than children of the same age in adoptive and biological 124 families. The present article extends this comparison to the early adolescence years. 125 Research comparing the social development of adoptees and children in 126 residential care is limited and mainly based on parents/caregivers’ and teachers’ reports. 127 Even though there is some statistical correspondence between teachers’ and peers’ 128 perceptions, peers are the ones who provide the most valid assessment of social status 129 (van den Berg, Lansu, & Cillessen, 2015). Peer relationships from the peers’ points of 130 view in Eastern European adoptees and children in residential care have not yet been 131 examined in detail. Our goal is to study in greater depth the social integration of these 132 children in the context where these interactions occur more frequently (i.e., school), and 133 with direct measures of their integration in the peer group. 134 The Present Study 135 The aim of this study is to explore the social relationships of international 136 adoptees (IA) and children in residential care (RC) in the school context, compared with 137 children who grow up in their biological family (comparison group, CG). Firstly, this 138 paper analyzes the sociometric status of IA, RC and CG children in their peer groups. 139 Given the number of studies that have reported significant relational difficulties in IA 140 and RC school-age children, we were especially interested in studying whether the 141 percentage of rejected children would be higher in IA and RC compared to CG. 142 Secondly, we explored the number of nominations that children expressed and received 143 from peers, their reciprocal friendship and enmity relationships and their perceptions 144 about the peers who would nominate them (i.e., sociometric perceptions). Thirdly, we 145 explored peer reports of likeability and social behaviours in our three groups. 146 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 7 Method 147 Participants 148 The participants in this study were 76 children aged between eight and fourteen 149 years (M = 10.78, SD = 1.38). These children belonged to one of three groups: 150 international adoptees from Russia in Spain (IA; n = 24), children living in residential 151 care in Spanish institutions (RC; n = 19) and a comparison sample of Spanish children 152 living with their birth families without experiences of early adversity (CG; n = 33). 153 The adoptive group was composed of children born in Russia and adopted by 154 Spanish families at an average age of 36 months (SD = 15). In this group, the 155 percentage of boys (79%) was greater than that of girls (21%), an imbalance typical in 156 adoptions from Russia (Pascual, 2000). All the children in this group had been 157 institutionalised in their country of origin (average 27 months, SD = 14). 158 The group of children in RC (58% girls, 42% boys) was composed of Spanish 159 children separated from their birth families due to severe experiences of maltreatment. 160 They arrived at the residential centres at an average age of 5.97 years (SD = 1.31). 161 Compared with Eastern European institutions (e.g. O’Connor et al., 2000; Rutter & the 162 English and Romanian Adoptees study team, 1998), the Spanish protection centres have 163 more favourable material conditions. Caregivers have professional qualifications and 164 the number of children and caregivers is about similar, with caregivers rotating in turns. 165 While in these centres, children attend the schools in the community. 166 The CG was composed of Spanish children (58% boys, 42% girls) from different 167 neighbourhoods and socio-economic levels in the same city where most of the adopted 168 and institutionalised children lived. The CG children were living with their birth 169 families with no contact whatsoever with child protection. 170 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 8 Permission to contact schools was given by the families of the adoptive and 171 community group, as well as by the child protection authority responsible for 172 institutional care. The classrooms of these 76 children were visited to collect the 173 sociometric data. In total, the 76 target children were evaluated in their 66 classrooms 174 from 52 different schools (some of the target children attended the same schools and, 175 more exceptionally, the same classroom). In order to obtain the sociometric data of the 176 target children, all their classmates (n = 1621) completed the sociometric 177 questionnaires. The average class size was 25 students (SD = 3). Average class sizes 178 were statistically similar in the three groups of children (IA: M = 25, SD = 3; RC: M = 179 26, SD = 3; CG: M = 25, SD = 4), F(2, 74) = 0.78, p = .462. 180 Measures 181 Sociometric Status and Indices of Peer Relationships. Each child in each 182 classroom answered a sociometric questionnaire with four questions in which they had 183 to mention the classmates they most liked to be with (positive nominations), the 184 classmates they least liked to be with (negative nominations), the classmates that they 185 believed liked to be with him/her (positive perceptions) and the classmates that they 186 believed did not like to be with him/her (negative perceptions). The number of 187 nominations in each question was unlimited. This sociometric information was collected 188 during school hours. Using the software program SOCIOMET (González & García189 Bacete, 2010) 12 indicators described in Table 1 were analysed. The sociometric status 190 considered are those proposed by Coie, Dodge and Coppotelli (1982). The sociometric 191 classification is based on an adjusted probability method, whose validity for the 192 identification of sociometric status in the classroom has been demonstrated in 193 comparison with other methods (García-Bacete & Cillessen, 2017). Children are 194 assigned to one sociometric status category based on the positive and negative 195 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 15 them) were not statistically significant, indicating that children from the three groups 344 perceived a similar number of peers who disliked being with them. 345 Peer Reports of Likeability and Social Behaviours 346 The peers of the target children at school rated their likeability, prosociality, 347 withdrawal and aggressiveness. Table 4 shows the group comparisons. IA and RC 348 obtained lower scores in likeability than CG (p = .002 and p = .019, respectively). IA 349 and RC also obtained lower scores than the CG in prosociality (p = .001 and p = .002, 350 respectively) and RC obtained higher scores than the CG in aggression (p < .001). The 351 differences in aggression between IA and CG tended to statistical significance (p = 352 .066). Non-significant differences between groups were observed in social withdrawal, 353 although the effect size of the differences was medium. 354 Discussion 355 This study advances our knowledge of the relationships that adopted and 356 institutionalised children maintain at school and provides evidence that children 357 exposed to early adversity tend to present difficulties in their peer relationships later in 358 the school years. In contrast with most existing research based on self-reports or 359 information from parents and teachers, these results add highly valuable information 360 directly provided by peers, including the study of sociometric status, sociometric 361 perceptions, friendship and peer reports of social behaviours. 362 Peer Relations of Internationally Adopted Children 363 Sociometric Status and Social Preference. The analysis of the sociometric 364 status showed a generally unfavourable situation for our IA group. Children in our 365 sample adopted internationally from Russia are more likely to be rejected by their 366 classmates. Specifically, 45.8% are categorised as rejected, a percentage much higher 367 than the 15% in the comparison sample and in the normative samples (Cillessen & 368 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 16 Bukowski, 2018). Our findings differ from the study by Stams et al. (2000), in which 369 internationally early-adopted girls were more popular and less rejected by their peers 370 than normative samples. This difference is probably related to the gender of the 371 children, their countries of origin, their pre-adoptive history and their ages at adoptive 372 placement. The study by Stams et al. (2000) included a higher proportion of adopted 373 girls than in our sample (in mixed classrooms girls from the normative population tend 374 to be over-represented in the popular category and boys over-represented in the rejected 375 category, Cillessen & Bukowski, 2018), as well as children adopted at earlier ages and 376 from countries such as Colombia or Korea, where the material conditions in institutions 377 are more favourable than in Eastern European countries (O’Connor et al., 2000). 378 Our findings also differ from the more positive portrait of adopted children’s 379 peer relationships described in the meta-analysis by DeLuca et al. (2018). The studies 380 included in this meta-analysis did not consider sociometric information based on direct 381 assessment by peers and this could explain the difference, together with the specific age 382 of our sample, as discussed below. 383 The study by Palacios et al. (2013) on the same children at an average age of 384 seven years reported a more favourable sociometric position of adopted children, with 385 only 7% rejected by peers. Moreover, in that study, the adoptees were less likely to be 386 ignored by their peers and had a high social impact in their group. The previous study 387 was based on teachers’ reports and research has shown that the attunement of teachers 388 to the social status of their students is moderate (van den Berg et al., 2015). Also, the 389 present study considers children in their late childhood or early adolescence, with 390 previous research indicating more difficulties with peers as children adopted from 391 Eastern Europe grow older (Julian & McCall, 2016), particularly during the early 392 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 17 adolescence years (Howard et al., 2004). No doubt, the 46% of rejection by peers 393 reported here is a worrisome finding for our adoption group. 394 Friendship Relationships. The analysis of friendship shows a more favourable 395 situation, since IA have an average of three reciprocal friends in the classroom. 396 Nevertheless, when analysing the closest circle of friends, only 58% of adoptees have 397 significant friends at school (reciprocal friends when the number of nominations is 398 restricted to three). Some of these findings are consistent with previous research based 399 on parents’ and teachers’ reports indicating that most internationally adopted children 400 have quality relationships with their peers. For example, parents have reported that 401 about 65% to 80% of post-institutionalised international adoptees have a best friend or a 402 small group of friends (Humphreys et al., 2018) and 95% of adoptees report having 403 friends (Cohen & Westhues, 1995; Hawk & McCall, 2014), similar to what sociometric 404 information shows in this study. However, the reduction in significant friendship 405 described here has not been reported beforehand. Together with the high level of 406 rejection by peers, this is a worrying finding. 407 Sociometric Perceptions. Adopted children perceive being rejected by an 408 average of two classmates, but they actually received an average of eight negative 409 nominations. This means that these children’s social perception is not entirely in line 410 with reality. Some studies have shown that IA children with a history of institutional 411 care tend to present more difficulties than normative samples in social information 412 processing and in interpreting social cues (e.g. Wismer Fries & Pollak, 2004; 413 Humphreys, 2018). These difficulties could explain the social misperceptions of the IA 414 children in this study. 415 Peers Reports of Likeability and Social Behaviours. Our findings indicate that 416 likeability in the group of peers is lower in IA than in the CG children. Additionally, 417 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 18 peers describe IA children as less prosocial and somewhat more aggressive than non418 adopted children. This is consistent with prior studies showing that peer acceptance is 419 poorer in international adoptees placed in their new families after their first months of 420 life (Pitula et al., 2014). Prior literature has also suggested that, compared with their 421 non-adopted peers, international adoptees are less prosocial and tend to share less when 422 playing with other children (Pitula, Depasquale, Mliner, & Gunnar, 2017; Pitula et al., 423 2014). 424 Peer Relationships of Children in Residential Care 425 Sociometric Status and Social Preference. Compared with the other two 426 groups in our study, the majority of children in RC occupy an intermediate sociometric 427 position, although the percentage of rejected children is also high (26%) compared to 428 the community group (15%). This sample of institutionalised children had been studied 429 previously at an average age of seven years using teachers’ reports (Palacios et al., 430 2013) and they already had more difficulties than normative samples in their integration 431 in the group of peers. Other studies have also suggested that institutionalised children 432 are more isolated (Martín & Muñoz de Bustillo, 2009; Vorria et al., 1998) and more 433 rejected by schoolmates to do academic tasks (Martín et al., 2008) than the community 434 children. In our study, compared with the IA group, the RC children were less likely to 435 be rejected and more likely to be average. It is interesting that the lower sociometric 436 status of IA children is not observed in our RC children. Since these children live in 437 group settings, perhaps they develop the “hyper-cooperativeness” described by Keil et 438 al. (2018) and this facilitates their integration in the peer group. Also, the institutional 439 circumstances of these children are much better (e.g., material conditions, caretakers’ 440 training, children-caretakers ratio, integration in the schools of the community) than the 441 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 19 Eastern European institutions described in other studies (e.g., O’Connor et al., 2000; 442 Rutter & the English and Romanian Adoptees study team, 1998). 443 Friendship Relationships. For RC children, this study reports that 90.5% have 444 reciprocal friends in the classroom, with an average of between three and four. Previous 445 studies, based on peers and self-reports, have reported higher percentages (from 50% to 446 66%) of institutionalised children with no reciprocal friends at school (Argumendo & 447 Albornoz, 2006; Martín et al., 2011). What our study adds is that when the closest circle 448 of three reciprocal nominations is considered, 31% of RC children have no significant 449 friends in the classroom. In general, our results suggest that children in residential care 450 have a network of reciprocal friends at school, although the size of this network is small 451 when compared with normative samples, in line with other studies (Bravo & Fernández 452 del Valle, 2003; Martín & Dávila, 2008). 453 Sociometric Perceptions. Children in residential care perceived a level of peer 454 acceptance similar to the community children in our sample. However, these children 455 were rejected by more peers than the ones they identified (on average, they received 456 seven negative nominations but were able to identify four). Like happened in the 457 adoptive group, the early maltreatment experiences of this group may have an impact on 458 their social information processing. For example, in emotion recognition tasks, victims 459 of abuse have greater difficulty interpreting facial expressions of joy or sadness (Gibb, 460 Schofield, & Coles, 2009), or neutral or friendly expressions (Leist & Dadds, 2009). 461 These results are also consistent with studies with normative samples suggesting direct 462 and indirect relations between peer rejection and social information processing biases 463 (e.g. Lansford, Malone, Dodge, Pettit, & Bates, 2010). As discussed above, it is possible 464 that these difficulties in social information processing interfere with their perception of 465 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 20 peers’ acceptance. If this is the case, the use of self-report measures of peer integration 466 should be questioned for these groups of youngsters. 467 Peers Reports of Likeability and Social Behaviours. Our results show that 468 peer ratings of likeability are significantly lower in institutionalised children than in the 469 normative sample. Additionally, classmates describe institutionalised children are less 470 prosocial and more aggressive than their peers. This information is consistent with the 471 existing literature showing that peers in the classroom, teachers and caregivers tend to 472 describe these children more negatively, with greater behavioural problems than their 473 classmates and less able to solve conflicts (Attar-Schwartz, 2009; Martín et al., 2008). 474 In summary, based on the information provided by peers, our findings show 475 significant differences in the peer relationships of children with early adversity 476 compared to those with more normative trajectories. In this comparison, adopted and 477 institutionalised children in our sample have a number of similarities: more negative 478 nominations by peers, more negative than positive evaluations, fewer reciprocal friends, 479 lower scores of likeability and prosociality. But the comparison also indicates that, on 480 average, in their early adolescence years, the adoption group is more rejected by peers 481 and has a lower percentage of significant friends in the classroom, while the 482 institutionalised group is rated by their peers as more aggressive. 483 Strengths, Limitations, Future Lines of Research and Practical Applications 484 The use of peers as informants of the social integration of adopted and 485 institutionalised children is uncommon in existing research. The rich information 486 obtained through this direct assessment in the school context is one of the strengths of 487 this article. However, access to these samples, as well as caregivers’ and teachers’ 488 participation, are not simple tasks. The sample size was small (even though sociometric 489 data was collected in 66 classrooms) and gender was not balanced in the adoptive 490 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 21 group, with the methodological consequences that this entails. Gender differences were 491 observed in aggression and prosocial behaviour, but could not be controlled in further 492 between-groups comparisons. Future studies with larger and more balanced samples 493 could address this issue. Another priority for future research is to go deeper in the study 494 of potential developmental cascades (Masten & Cicchetti, 2010) that could explain the 495 peer integration difficulties experienced by these children, taking into account factors 496 such as neurobiological risks, executive function or indiscriminate social engagement, 497 among other things. Future studies with larger samples could also consider nested 498 analyses to explore the effect of other contextual variables (e.g., classroom climate) on 499 children’s social integration. 500 This research provides key information for the intervention with adopted and 501 institutionalised children. A proper identification of isolation or rejection by peers is a 502 first necessary step, and for this the use of sociometric techniques seems advisable. 503 Once identified, actions to improve these children’s social skills, to promote their 504 integration into their peer group and to avoid unfavourable circumstances (isolation, 505 rejection) should be undertaken, coordinating efforts between families, protection 506 centres and schools. Very often, the educational community is uninformed about the 507 specific needs of children with early adversities. When this happens, it is unlikely that 508 teachers are aware of the difficulties they may experience in the school context, which 509 highlights the importance of developing strategies to strengthen cooperation between all 510 those involved in promoting these children’s personal and social development. 511 Conclusions 512 In their late childhood and early adolescent years, children adopted from Russia 513 and those living in Spanish child protection centres show considerable difficulties in 514 their relationships with peers in the school context. Similar in a number of problems, 515 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 22 they also show specific difficulties. Awareness of their problems and integrated work 516 between families, protection centres and schools are necessary in the promotion of the 517 social development of children like those studied herein. Child protection and education 518 professionals are key to ensure that this happens. 519 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 23 References 520 Argumendo, D., & Albornoz, C. (2006). Calidad de la amistad en niños de 8 a 10 años 521 de una institución pública [Quality of friendship in children aged eight to ten years 522 from a public institution]. Revista de Psicología de La PUCP, 24(1), 83–107. 523 Attar-Schwartz, S. (2009). School functioning of children in residential care: The 524 contributions of multilevel correlates. Child Abuse and Neglect, 33(7), 429–440. 525 doi:10.1016/j.chiabu.2008.12.010 526 Barcons-Castel, N., Fornieles-Deu, A., & Costas-Moragas, C. (2011). International 527 adoption: Assessment of adaptive and maladaptive behavior of adopted minors in 528 Spain. The Spanish Journal of Psychology, 14(1), 123–132. 529 doi:10.5209/rev_SJOP.2011.v14.n1.10 530 Barone, L., Lionetti, F., & Green, J. (2017). A matter of attachment? How adoptive 531 parents foster post-institutionalized children’s social and emotional adjustment. 532 Attachment and Human Development, 19(4), 323-339. 533 doi:10.1080/14616734.2017.1306714 534 Bravo, A., & Fernández del Valle, J. (2003). Las redes de apoyo social de los 535 adolescentes acogidos en residencias de protección. Un análisis comparativo con 536 población normativa [The social support networks of adolescents in protection 537 centres. A comparative analysis with normative population]. Psicothema, 15(1), 538 136–142. 539 Caprin, C., Benedan, L., Ballarin, L., & Gallace, A. (2017). Social competence in 540 Russian post-institutionalized children: A comparison of adopted and non-adopted 541 children. Children and Youth Services Review, 75(1), 61–68. 542 doi:10.1016/j.childyouth.2017.02.020 543 Christoffersen, M. N. (2012). A study of adopted children, their environment, and 544 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 24 development: A systematic review. Adoption Quarterly, 15(3), 220–237. 545 doi:10.1080/10926755.2012.700002 546 Cillessen, A. H., & Bukowski, W. M. (2018). Sociometric Perspectives. In W. M. 547 Bukowski, B. Laursen, & K. H. Rubin (Eds.), Handbook of peer interactions, 548 relationships, and groups (2nd ed., pp. 64-83). New York, NY: Guilford 549 Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd Ed.). New 550 York Lawrence Erlbaum Associates 551 Cohen, J., & Westhues, A. (1995). A comparison of self-esteem, school achievement, 552 and friends between intercountry adoptees and their siblings. Early Child 553 Development and Care, 106(1), 205–224. doi:10.1080/0300443951060116 554 Coie, J. D., Dodge, K. A., & Coppotelli, H. (1982). Dimensions and types of social 555 status: A cross-age perspective. Developmental Psychology, 18(4), 557–570. 556 doi:10.1037/0012-1649.18.4.557 557 DeLuca, H. K., Claxton, S. E., & van Dulmen, M. H. (2018). The peer relationships of 558 those who have experienced adoption or foster care: A meta-analysis. Journal of 559 Research on Adolescence, 29(4), 796-813. doi:10.1111/jora.12421 560 Department for Education (2018). Children looked after in England (including 561 adoption), year ending 31 March 2018. Retrieved from 562 https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attach 563 ment_data/file/757922/Children_looked_after_in_England_2018_Text_revised.pdf 564 García-Bacete, F. J., & Cillessen, A. H. (2017). An adjusted probability method for the 565 identification of sociometric status in classrooms. Frontiers in Psychology, 8(1), 1– 566 12. doi:10.3389/fpsyg.2017.01836 567 Garcia-Quiroga, M., Hamilton-Giachritsis, C., & Ibañez, M. (2017). Attachment 568 representations and socio-emotional difficulties in alternative care: A comparison 569 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 31 Table 1 721 Indices of peer relationships obtained from the sociometric assessment 722 Indexes Description Nominations expressed by the target child Positive nominations Number of peers that the child likes being with Negative nominations Number of peers that the child dislikes being with Positive perceptions Number of peers that the child believes like to be with him/her Negative perceptions Number of peers that the child believes do not like to be with him/her Nominations received by the target child Positive nominations Number of peers who like being with the child Negative nominations Number of peers who do not like being with the child Composite scores Reciprocal friendships Number of peers who like being with the child and the child also likes being with them Reciprocal enmities Number of peers who do not like being with the child and the child does not like being with them either Significant friendships Number of reciprocal friendships when only the first three nominations of each child are considered Social preference Percentage of the difference between the number of peers who like being with the child and the number of peers who do not like being with the child Opposing feelings Number of cases in which the child positively nominates a peer and is negatively nominated by the latter, or the other way around Sociometric status Sociometric status (based on Coie, Dodge & Coppotelli, 1982): preferred, neglected, rejected, controversial, average 723 724 725 726 727 728 729 730 731 732 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 32 Table 2 733 Sociometric status in adopted children, children in residential care and the community 734 group 735 Adopted group (n = 24) Residential care group (n = 19) Community group (n = 33) Preferred 2 (8.3%) 1 (5.3%) 5 (15.2%) Average 9 (37.5%) 11 (57.9%) 22 (66.7%) Neglected 1 (4.2%) 1 (5.3%) 0 (0%) Rejected 11 (45.8%) 5 (26.3%) 5 (15.2%) Controversial 1 (4.2%) 1 (5.3%) 1 (3%) 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 33 Table 3 760 Average scores of the three groups of children in each index and comparisons between 761 the groups 762 Measure Group n M SD Welch’s F Effect size a Pairwise comp. Post hoc p values Positive nominations expressed CG 32 7.94 2.18 2.02 .049 CG-IA .141 IA 24 6.54 2.98 CG-RC .556 RC 19 7.05 3.29 IA-RC .859 Negative nominations expressed CG 32 3.97 2.47 0.47 .017 CG-IA .999 IA 24 3.96 2.91 CG-RC .622 RC 19 4.84 3.59 IA-RC .663 Positive nominations received CG 33 8.45 4.25 4.35* .107 CG-IA .014 IA 24 5.58 3.20 CG-RC .141 RC 19 6.21 3.88 IA-RC .838 Negative nominations received CG 33 3.27 3.73 10.86*** .229 CG-IA .000 IA 24 8.25 4.80 CG-RC .007 RC 19 7.05 4.24 IA-RC .664 Social preference CG 33 25.12 32.69 10.01*** .225 CG-IA .000 IA 24 -11.25 31.24 CG-RC .007 RC 19 -2.32 27.68 IA-RC .586 Reciprocal friendships CG 32 5.44 2.74 6.39** .158 CG-IA .003 IA 24 3.17 2.18 CG-RC .033 RC 19 3.58 2.29 IA-RC .822 Reciprocal enmities CG 32 0.75 1.22 5.00* .101 CG-IA .078 IA 24 2.00 2.50 CG-RC .035 RC 19 1.95 1.75 IA-RC .996 Opposing feelings CG 32 1.22 1.16 2.02 .049 CG-IA .232 IA 24 1.92 1.79 CG-RC .292 RC 19 1.84 1.54 IA-RC .988 Positive perceptions expressed CG 32 5.16 3.23 3.43* .068 CG-IA .082 IA 24 3.58 2.13 CG-RC .972 RC 19 5.37 3.25 IA-RC .114 Negative perceptions expressed CG 32 3.41 2.47 2.67 .073 CG-IA .261 IA 24 2.42 2.19 CG-RC .555 RC 19 4.32 3.28 IA-RC .093 Note: CG = community group; IA = adoptive group; RC = Residential care group. 763 a Effect sizes = η2p (.01 small, .06 medium, .14 large). 764 *p < .05, **p < .01, ***p < .001. 765 766 767 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 34 Table 4 768 Peer ratings of likeability, prosociality, aggression and withdrawal for each group of 769 children. 770 Measure Group n M SD Welch’s F Effect size a Pairwise comp. Post hoc p values Likeability CG 28 .517 0.92 7.35** .187 CG-IA .002 IA 24 -.432 0.97 CG-RC .019 RC 18 -.227 0.83 IA-RC .741 Prosociality CG 28 .588 1.05 8.95*** .234 CG-IA .001 IA 24 -.403 0.78 CG-RC .002 RC 18 -.378 0.74 IA-RC .994 Withdrawal CG 28 -.149 0.87 2.75 .077 CG-IA .105 IA 24 .375 0.94 CG-RC .927 RC 18 -.268 1.17 IA-RC .151 Aggression CG 28 -.494 0.91 9.82*** .200 CG-IA .066 IA 24 .122 1.01 CG-RC .000 RC 18 .606 0.76 IA-RC .190 Note: CG = community group; IA = adoptive group; RC = Residential care group. 771 a Effect sizes = η2p (.01 small, .06 medium, .14 large). 772 *p < .05, **p < .01, ***p < .001. 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 PEER RELATIONSHIPS, ADOPTION AND RESIDENTIAL CARE 35 790 Figure 1. Percentage of children in each group with zero, one, two or three significant 791 friends in the classroom. 792 12,5 41,7 31,6 21,9 25 31,6 37,5 29,2 31,6 28,1 4,2 5,3 0% 20% 40% 60% 80% 100% Community group Adoption group Residential care group No significant friends 1 significant friend 2 significant friends 3 significant friends