Intergroup attitudes and contact between Spanish and immigrant-background adolescents using network analysis.

This study aimed to analyze the relationship among different evaluative reactions of the intergroup attitudes and contact in Spanish adolescents evaluating different ethnic minorities and in immigrant-background adolescents evaluating Spanish youth. This study was based on psychosocial models of great impact in the study of intergroup relations such as the Stereotype Content Model and the Behaviors from Intergroup Affect and Stereotypes Map, and incorporated a new approach to the study of attitudes: psychological networks. In total, 1122 Spanish adolescents and 683 adolescents with an immigrant background (Moroccan, Romanian or Ecuadorian origin) participated in the study, aged from 12 to 19 years. They answered a questionnaire with measures of stereotype dimensions (morality, immorality, sociability and competence), emotions (positives and negative), behavioral tendencies (facilitation and harm) and contact (quantity and quality). The results show similar structural patterns in the six studied groups, with emotions acting as links between stereotypes and behavioral tendencies. Moreover, positive and negative stereotype dimensions appeared as independent dimensions that were part of different processes: sociability and morality, and competence to a lesser extent, were related to facilitation behaviors through positive emotions, while immorality was related to harm behaviors through negative emotions. This could indicate that, to achieve successful intergroup relations involving cooperation and the development of friendly relationships, it would be appropriate to intervene in parallel in these two pathways. Due to the centrality of positive emotions (and sociability and immorality) and, therefore, their capacity to affect the entire network, focusing interventions on these variables could be an appropriate strategy to achieve overall positive attitudes.

Introduction

The extensive literature of intergroup attitudes shows the importance of its study in contemporary societies due to the great impact that these attitudes have on intergroup relations. The increase in diversity in current societies is reflected in the presence of students with an immigrant background in secondary schools (in Spain, about 10% of the total student body during the 2019–2020 academic year whereas this percentage was less than 2% in the 2000–2001 academic year, according to the Spanish Ministry of Education, Culture and Sport [1, 2]). These spaces, which are the main contexts of socialization for adolescents, are also the main scenarios where intergroup contact takes place. In addition, adolescence is a key developmental stage in the formation of intergroup attitudes, in which interventions to improve them can be particularly effective [3].

In the present study, we analyzed the structure of intergroup attitudes of different groups of adolescents living in Spain from both the majority and minority perspective, and their relationship with the quality and quantity of contact. For this purpose, this study was based on psychosocial models of great impact in the study of intergroup relations such as the Stereotype Content Model (SCM) [4] and the Behaviors from Intergroup Affect and Stereotypes Map (BIAS Map) [5, 6], and incorporated a new approach to the study of attitudes: psychological networks [7].

Current perspectives in the study of intergroup attitudes

Many conceptualizations have described attitudes as evaluations about attitudinal objects (people, groups, things, ideas, etc.), ranging from positive to negative, that can be inferred through beliefs, affect and behavior [8, 9]. According to SCM [4], group representations (stereotypes) are structured in two fundamental evaluative dimensions: warmth and competence. The first one refers to the perceived intentions of the out-group or their members with aspects related to kindness and goodness (e.g., whether they are sincere, affectionate), and the second one refers to the capacity of out-group or their members to achieve their objectives or intentions (e.g., whether they are intelligent, efficient). This model also shows that different combinations in the warmth and competence dimensions cause different patterns of affective reactions (i.e., contempt, admiration, compassion and envy).

Subsequent studies have shown three stereotype dimensions, dividing in two the dimension of warmth: morality and sociability [10, 11]. The morality dimension would include characteristics related to the correctness of out-group targets (e.g., honest, trustworthy), while sociability would include characteristics related to cooperation and connection with others (e.g., friendly, kind [12]). Other authors [13, 14] have included the negative pole of morality (i.e., immorality) and have shown its importance in out-group evaluations. Previous research [13-17] has shown evidence of the theoretical importance and the diagnostic capacity of the negative pole of the morality domain (i.e., immorality) in the out-group’s impression formation (i.e., for this dimension, negative information is more enlightening than positive information).

The extension of SCM [4], i.e. the BIAS Map [5, 6], provides a general structure about how the stereotypes and emotions experienced towards out-groups are related to behaviors towards them. Intergroup emotions are those that people experience due to belonging to a group and that result from self-categorization and salient social identity [18]. Studies based on the BIAS Map [5, 6] as well as in the Intergroup Emotion Theory (IET; [19]) have showed the powerful effects of intergroup emotions acting as mediators between stereotypes and behavioral tendencies. The meta-analysis of Talaska et al. [20] showed that emotions towards ethnic minorities were more direct predictors of discriminatory behaviors than beliefs and stereotypes. The IET model, however, qualifies that the sequence cognition, emotion and behavior is not always unidirectional, because emotion has broad forward and backward effects on cognition (i.e., “appraisals” and emotions influence each other) [19].

The BIAS Map hypothesis have been tested in different cultural contexts. In Norway, Bye and Herrebrøden [21] found support general for the BIAS Map predictions, with warmth and competence as stereotype dimensions, but with some variations. In Spain, Cuadrado et al. [22] showed the relationship between stereotypes (morality, sociability and competence dimensions) and behavioral tendencies of facilitation through positive emotions in adults and adolescents, from the perspective of the majority and minority. SCM extended (three stereotype dimensions) and BIAS Map have also been used as a theoretical framework for the development of two studies on the adolescent population in Spain [23, 24]. In the first study [23], morality predicted facilitation and harm intentions (through positive and negative emotions) and harm intentions (through negative emotions), while sociability predicted only facilitation intentions through positive emotions. In the second study [24], morality predicted the intentions of facilitation and harm through emotions of admiration and contempt, respectively. In general, morality seems to be a key determinant in predicting intergroup behavior through emotions. Some of the above findings have been captured in the three key assumptions of the Moral Primacy Model [25]: 1) Morality, sociability, and competence are distinct dimensions that make unique contributions to impression formation; 2) Morality has a primary role in the formation of impressions and the evaluations we make about other people and groups both at different stages of impression formation and in behavioral outcomes; 3) The primary role of morality is due to its close link to the judgment of whether other social targets represent an opportunity or a threat.

Some researchers consider both positive and negative attitudes to be separate dimensions and point to the appropriateness of including both dimensions in the study of attitudes [26, 27]. Pittinsky et al. [28] showed that positive and negative attitudes are functionally separable since they relate to behaviors according to their valence (i.e., positive attitudes are more strongly linked to positive behaviors, while negative attitudes are more strongly linked to negative behaviors). Therefore, improving intergroup relations may involve both the reduction of negative intergroup attitudes and the promotion of positive intergroup attitudes, because this reduction and promotion may be linked to different psychological processes [27].

Consistent with the SCM [4] and the BIAS Map [5, 6], stereotypes, emotions and behavioral tendencies were used as indicators of intergroup attitudes in this study. The present study also includes the stereotype dimension of immorality in order to test whether the positive (i.e., morality, sociability or competence) and negative (i.e., immorality) poles are related to positive and negative emotions and to positive (facilitation) and negative (harm) behaviors, respectively.

The study of attitudes from a network approach

Of the many frameworks used in the study of attitudes, recently, a new approach is gaining significant interest: the psychological networks [7, 29–31]. This approach has led to a new way of looking at the concept of attitude and a new measurement model to apply to empirical data [7]. Traditionally, for the study of psychological constructs, it is assumed that there are underlying latent variables that explain the scores of the observable variables (e.g., self-reported items) and the correlations between these observable variables. From a network approach, instead of considering observable variables as indicators of a latent variable, they are considered as autonomous entities related in a network of dynamic systems [32]. This allows us to overcome some of the conceptual and methodological limitations or problems found in research based on latent variable models (and derived from the tripartite attitude model in attitudes research; for a more detailed view see [7, 31]). In addition, networks improve other methodological approaches such as the multiple regression model or factor analysis, fundamentally because it shows in a single plot the relationships between variables of large datasets, building a network and providing an easier interpretation [33].

Within this framework, a new measurement model has been developed to explain how the different information bases of attitude relate to the attitude construct and to explain correlations between them: the Causal Attitude Network (CAN) model [7]. This model conceptualizes attitudes as networks consisting of evaluative reactions (beliefs, feelings and behaviors) toward the attitude object and interactions between these reactions. The present study followed this line and, therefore, no directionality was established between the different evaluative responses, to be compatible with the bidirectionality between cognition and emotions defended by some theories of emotions such as the IET [19] and found in some studies in recent research [22].

Intergroup attitudes and contact

One of the most important approaches to improving intergroup relations is intergroup contact theory, based on the contact hypothesis developed by Allport [34] (see also the meta-analysis of Pettigrew and Tropp, [35]). This hypothesis proposes that intergroup contact could decrease prejudice and improve intergroup relations between members of majority and minority groups (under appropriate conditions: equal status between the groups, common goals, intergroup cooperation, and the support of authorities, law, or custom). However, in Pettigrew and Tropp’s meta-analysis, it was observed that these conditions were not necessary for a reduction in prejudice to occur and these authors proposed that the simple familiarity that occurs with contact generates liking. In addition, they point out that the relationship between familiarity and liking could be influenced by variables such as the reduction of uncertainty, intergroup anxiety and threat. The study by Laurence et al. [36] showed that increased out-group size was related to intergroup attitudes towards them only in more segregated communities (but not in the case of more integrated communities), and that this is explained by an increase in perceived threat and a decrease in the quality of contact (when the increase in out-group size exceeds a certain value).

Specifically, it is the quality of contact rather than the quantity of contact that has shown the greatest effect on intergroup attitudes [37]. Quality of contact refers to the participant’s subjective sense of the nature of the contact experience [38]. In a study by Binder et al. [39], both the quality and quantity of contact influenced over time two indicators of prejudice: the desire for social distance and negative intergroup emotions, with the quality of contact being the most relevant variable. Some studies have shown that positive contact, conceptualized as cross-group friendship, is related to different measures of intergroup attitudes, especially emotions like admiration and sympathy [37, 38]. Bobowik et al. [40] conducted a study using a social network approach in Spain with adults of immigrant background and found that the proportion of intergroup contacts (with native-born population) representing strong ties (i.e., close relationships such as friends, romantic partners, and immediate and extended family members) and ethnic diversity among these ties were associated with more favorable attitudes toward outgroups. In another study conducted in the same community with 15- and 16-year-old students, it was observed that students who reported frequent interaction with other students of different nationalities and religions held more positive intergroup attitudes and the presence of a greater number of immigrants in classrooms was related to lower levels of xenophobia [41].

The present study

This study aimed to examine the relationship between different evaluative reactions of the attitude (stereotypes, emotions, behavioral tendencies) and contact in different groups of adolescents in Spain: in the majority group (Spanish adolescents evaluating immigrant youths from different origins) and in three minority groups with an immigrant background (adolescents evaluating Spanish youth from three of the main immigrant groups in Spain according to the Spanish National Institute of Statistics [42]: Moroccans, Romanians and Ecuadorians). The application of network models in different groups and/or towards different targets will allow us to explore the structure and dynamic properties of intergroup attitudes (i.e., the relations between different evaluative reactions). As far as we know, the present study is one of few studies that simultaneously included the positive and negative pole of stereotypes, emotions and behavioral tendencies applied to the adolescent population (majority and different ethnocultural minorities) using a network approach.

Firstly, we think morality, and especially immorality, are probably the most influential stereotypical dimensions in the network (Hypothesis 1). We also expect to find a central role for emotions within the network since the literature has shown their powerful effect acting as the link between all the other information bases of the attitude in adults [5, 6, 19, 22] and adolescent population [23, 24] (Hypothesis 2). Thirdly, we also hypothesize that the relationship between stereotypes and behavior through emotions is a function of their valence (Hypothesis 3). That is, we expect to find a relationship of stereotypes to facilitation tendencies through positive emotions and a relationship of stereotypes to harm tendencies through negative emotions [26-28]. Finally, we presume that the quality of contact has more influence on the network than the quantity of the contact, and that it is related to the rest of the network, mainly through emotions [37, 38] (Hypothesis 4).

Materials and methods

Participants

A sample of 1122 Spanish adolescents and 683 adolescents with an immigrant background from 12 to 19 years voluntarily participated in this study. From the sample of Spanish adolescents, 488 evaluated youth of Moroccan origin (SM group: M = 14.79, SD = 1.23; 52.4% girls), 314 evaluated youth of Romanian origin (SR group: M = 14.84, SD = 1.53; 54.2% girls) and 320 evaluated youth of Ecuadorian origin (SE group: M = 15.36, SD = 1.57; 51.7% girls). From the sample of adolescents with an immigrant background, 360 were of Moroccan origin (M group: M = 15.16, SD = 1.36; 58.7% girls), 137 were of Romanian origin (R group: M = 15.04, SD = 1.45; 56.6% girls) and 186 were of Ecuadorian origin (E group: M = 15.27, SD = 1.42; 48.91% girls). All of them evaluated Spanish youth. They were enrolled in 17 public secondary schools in different municipalities in five Spanish provinces (Almería, Alicante, Castellón, Madrid and Murcia). The valid percentages of first-generation adolescents were 63.2%, 68.9% and 41.6% for adolescents of Moroccan, Ecuadorian and Romanian origin, respectively. For second generation, the valid percentages were 36.8%, 31.1% and 58.4% for adolescents of Moroccan, Ecuadorian and Romanian origin, respectively. The mean age of arrival in Spain was 4.82 years old (SD = 4.39) for Moroccan origin adolescents, 5.49 (SD = 4.35) for Romanian origin adolescents, and 6.38 (SD = 4.83) for Ecuadorian origin adolescents. Average perceived socioeconomic status score (ranging from 1 to 10 points) was 6.38 (SD = 1.36) for SM group, 6.39 (SD = 1.29) for SR group, 6.43 (SD = 1.33) for SE group, 6.14 (SD = 1.61) for M group, 6.36 (SD = 1.28) for R group, and 6.69 (SD = 1.36) for E group.

Variables and instruments

Participants answered a questionnaire with two similar versions, changing the out-group evaluated. Spanish adolescents answered the questionnaire assessing Moroccan youth, Romanian youth or Ecuadorian youth, and adolescents with an immigrant background assessed Spanish youth. Questionnaires contained instruments to measure the following variables. The estimated reliability coefficients (Cronbach’s alpha) are included in Table 1.

Table 1

Coefficients of estimated reliability (Cronbach’s alpha and Split-half after Spearman-Brown correction) of all the variables in the six sub-samples.

	SM group		SR group		SE group		M group		R group		E group
Variables	α	Split-half	α	Split-half	α	Split-half	α	Split-half	α	Split-half	α	Split-half
Morality	.82	.84	.79	.80	.80	.80	.65	.65	.64	.69	.66	.72
Immorality	.80	.74	.86	.84	.84	.81	.77	.74	.82	.81	.80	.78
Sociability	.85	.86	.85	.87	.81	.80	.72	.71	.75	.80	.82	.80
Competence	.76	.76	.77	.77	.78	.76	.68	.63	.69	.66	.69	.67
Positive emotions	.86	.86	.85	.86	.87	.87	.77	.79	.80	.77	.81	.83
Negative emotions	.83	.81	.86	.86	.79	.75	.72	.71	.81	.79	.81	.77
Facilitation	.91	.89	.91	.87	.93	.91	.84	.77	.85	.82	.85	.79
Harm	.79	.82	.85	.88	.72	.77	.67	.67	.77	.85	.60	.72
Quantity of contact	-	-	-	-	-	-	-	-	-	-	-	-
Quality of contact	.79	-	.73	-	.82	-	.73	-	.74	-	.81	-

SM group = Spanish adolescents who evaluate Moroccan-origin youth; SR group = Spanish adolescents who evaluate Romanian-origin youth; SE group = Spanish adolescents who evaluate Ecuadorian-origin youth; M group = Moroccan-origin adolescents; R group = Romanian-origin adolescents; E group = Ecuadorian-origin adolescents.

Stereotypes

This variable was measured through a 17-item scale elaborated from López-Rodríguez et al. [43] and Sayans-Jiménez et al. [14]. The scale consisted of four subdimensions: morality (4 items: honest, trustworthy, sincere and respectful), immorality (5 items: aggressive, malicious, harmful, treacherous and false), sociability (4 items: friendly, warm, likeable and kind) and competence (4 items: intelligent, skillful, competent, and efficient). Participants were asked to what extent each adjective described to out-group youth (Spaniards, Moroccans, Romanians or Ecuadorians). The response-scale ranged from 1 (not at all) to 5 (very much). Average scores were calculated (ranging from 1 to 5), so that higher scores indicated greater perception of morality, immorality, sociability and competence of the out-group evaluated.

Intergroup emotions

This variable was measured through a 13-item scale elaborated from Cuadrado et al. [44] and the subscale of emotions of the Prejudiced Attitude Test [45]. The scale consisted of two subdimensions: positive emotions (admiration, respect, security, understanding, love and sympathy) and negative emotions (mistrust, indifference, hate, anger, fear, disgust and discomfort). Participants were asked to what extent they feel each emotion toward the out-group. The response-scale ranged from 1 (nothing) to 5 (a lot). Average scores were calculated (ranging from 1 to 5), so that higher scores indicated more intense emotions (positive or negative) toward the out-group evaluated.

Behavioral tendencies

This variable was measured through a 12-item scale from López-Rodríguez et al. [46], based on the BIAS Map [5, 6]. The scale consisted of two subdimensions: facilitation (caring for them, helping them, protecting them, cooperating with them, teaming up with them, joining them) and harm (excluding them, ignoring them, going beyond them, insulting them, assaulting them, harassing them). Participants were asked to what extent they would be willing to do each behavior toward out-group people. The response-scale ranged from 1 (nothing) to 5 (a lot). Average scores were calculated (ranging from 1 to 5), so higher scores indicated greater willingness to engage in facilitation and harm behaviors towards the out-group evaluated.

Quantity of contact

A question (In general, how much contact have you had or do you have with -the out-group-?) was used as an indicator of the amount of contact. The response-scale ranged from 1 (nothing) to 5 (a lot). Higher scores indicated a higher quantity of contact.

Quality of contact

This was measured using a 3-item scale based on Islam and Hewstone [47]. The participants were asked about the quality of the contact with the out-group by using a list of bipolar adjectives with three response alternatives ranging from unpleasant-pleasant, involuntary-voluntary and superficial-intimate. Average scores were calculated (ranging from 1 to 5), so higher scores indicated better quality of contact.

Socio-demographic variables

Participants reported their sex, age, place of birth, age when they arrived to Spain, countries of birth of their parents and perceived socioeconomic status.

Procedure

After carrying out a residential census analysis, five sampling zones (Almería, Alicante, Castellón, Madrid and Murcia) were chosen with the largest immigrant population [34]. Specifically, the percentages of foreign population in these provinces were 21.37%, 19.84%, 15.18%, 14.15% and 14.64%, respectively according to the National Institute of Statistics [48]. Members of the research team contacted the public secondary schools in these areas to choose those with the greatest proportion of immigrant students. All the permits required from the local governments and schools were obtained. Once the school authorities agreed to participate, the adolescents’ parents were informed of the relevant aspects of the research (voluntary and anonymous participation, use of the data for scientific purposes, etc.) and they signed a consent form. The questionnaires were administered by members of the research team in the classrooms, in paper-and-pencil format. The duration was approximately 30 minutes. The study was approved by the Bioethics Committee in Human Research of the University of the authors. The database used in this research has been made publicly available and can be accessed at Open Science Framework (OSF): https://osf.io/reyh3/?view_only=f7bf63a972c14a4bb0bfe97c6c05e7c9.

Results

Data analysis

We computed the descriptive statistics, partial correlations and Cronbach alpha coefficients for each scale and group using IBM SPSS Statistics v.25. We estimated an undirected psychological network where each node represents one of the ten domains measured: four stereotype dimensions (morality, immorality, sociability and competence), two intergroup emotions (positive and negative emotions), two behavioral tendencies (facilitation and harm) and quantity of contact and quality of contact. We used these 10 domain scores as variables in the Gaussian Graphical Model (GGM; [49]), a regularized partial correlation network [50]. Extended Bayesian Information Criterion function (EBICglasso) was used as an estimator. In the Gaussian graphical model, the partial correlation coefficients were directly used as edge weights between every two nodes in the network; there was no edge if the partial correlation coefficient was 0, e.g., Epskamp et al. [51]. The thickness of edges represents the strength of the relationships between variables; the absence of a line implies no (or a very weak) relationship between the variables. Due to the presence of positive and negative edges, the expected influence centrality index was estimated for all variables. This index provides specific information regarding the impact of each node on the rest of the network [52]: “aim to assess the nature and strength of a node’s cumulative influence within the network, and thus the role it may be expected to play in the activation, persistence and remission of the network” (p. 748). In addition, the closeness (the more a node is positioned to the center of the network, the closer it is to all other nodes) and betweenness (number of shortest paths passing through a node or the power of a node to interrupt the flow of information on the network) centrality indices were calculated. All indices are presented in a standardized way. JASP 0.14 software was used to carry out the analysis for network estimation and visualization, with the auto correlation method (automatically detects the type of input variable and uses the most suitable type of correlation).

Preliminary analysis

Reliability estimations are presented in Table 1, descriptive statistics in Table 2 and partial correlations in Table 3. The estimation of the reliability of the scores on each scale was adequate to good.

Table 2

Descriptive statistics of all the variables in the six sub-samples.

	SM group	SR group	SE group	M group	R group	E group
Variables	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)	M (SD)
Morality	2.76 (0.79)	3.42 (0.74)	3.45 (0.73)	3.29 (0.70)	3.38 (0.58)	3.06 (0.59)
Immorality	2.98 (0.79)	2.59 (0.82)	2.29 (0.76)	2.60 (0.82)	2.64 (0.78)	2.75 (0.75)
Sociability	3.04 (0.85)	3.64 (0.76)	3.70 (0.77)	3.66 (0.74)	3.94 (0.58)	3.70 (0.72)
Competence	3.26 (0.74)	3.62 (0.67)	3.45 (0.72)	3.50 (0.65)	3.53 (0.58)	3.56 (0.57)
Positive emotions	2.85 (0.85)	3.45 (0.80)	3.32 (0.85)	3.45 (0.70)	3.57 (0.66)	3.30 (0.70)
Negative emotions	2.41 (0.86)	2.04 (0.78)	1.87 (0.66)	2.17 (0.65)	2.10 (0.68)	2.16 (0.72)
Facilitation	3.30 (0.92)	3.91 (0.81)	3.77 (0.87)	3.94 (0.78)	4.11 (0.66)	3.77 (0.73)
Harm	1.51 (0.63)	1.34 (0.57)	1.30 (0.43)	1.56 (0.78)	1.44 (0.53)	1.35 (0.41)
Quantity of contact	3.59 (1.09)	3.91 (0.90)	3.67 (0.96)	4.15 (0.99)	4.50 (0.89)	3.77 (1.03)
Quality of contact	3.50 (0.76)	3.92 (0.72)	3.85 (0.77)	3.85 (0.76)	4.10 (0.72)	3.84 (0.77)

SM group = Spanish adolescents who evaluate Moroccan-origin youth; SR group = Spanish adolescents who evaluate Romanian-origin youth; SE group = Spanish adolescents who evaluate Ecuadorian-origin youth; M group = Moroccan-origin adolescents; R group = Romanian-origin adolescents; E group = Ecuadorian-origin adolescents.

Table 3

Partial correlations between all the variables in the six sub-samples.

	SM group	SR group	SE group	M group	R group	E group
M-I	-.25***	-.32***	-.29***	-.27***	-.32***	-.34***
M-S	.47***	.33***	.38***	.20**	.36***	.29***
M-C	.18***	.23***	.25***	.20***	.32***	.10
M-PE	.09*	.05	.16**	.12*	-.07	.18*
M-NE	-.09	.06	-.07	-.07	-.03	-.00
M-F	.10*	.08	.02	.05	.04	-.01
M-H	.07	.05	.14*	.13*	.14	.12
M-QC 1	-.03	-.03	.01	-.06	.01	-.01
M-QC 2	-.06	.14*	-.02	.13*	.03	.08
I-S	-.10*	-.18**	-.06	-.17**	-.05	.05
I-C	-.00	-.03	.08	-.00	-.06	.15
I-PE	-.04	-.12	-.09	.08	-.14	-.03
I-NE	.25***	.32***	.28***	.31***	.41***	.38***
I-F	.13**	.02	.04	.04	.03	.04
I-H	.12*	.06	.04	.09	.02	.06
I- QC 1	.03	.26***	.11	.00	-.10	.16*
I- QC 2	-.04	-.01	-.03	.04	.09	-.10
S-C	.17***	.26***	.23***	.23***	.21*	.06
S-PE	.25***	.25***	.12*	.26***	.28**	.13
S-NE	.05	-.01	-.16**	.06	-.04	-.17*
S-F	.05	.06	.00	.03	.01	.29***
S-H	.06	-.06	-.04	.01	-.04	.16*
S- QC 1	.09	.07	.02	.12*	.03	.01
S-QC 2	.18***	-.02	.11	.06	.08	.16*
C-PE	.07	.03	.12*	.21***	.09	.17*
C-NE	.06	.05	.11	-.06	.15	.02
C-F	.10*	-.00	-.02	.01	-.03	.15*
C-H	-.14**	-.10	-.11	-.01	-.18	.04
C-QC 1	-.08	.08	.04	-.06	-.09	.06
C-QC 2	-.01	.00	.02	-.07	.07	.01
PE-NE	-.14**	-.09	.07	-.12*	.14	.06
PE-F	.41***	.48***	.55***	.33***	.50***	.33***
PE-H	.09	.23***	-.08	.01	.13	-.04
PE-QC 1	.03	.05	.00	.04	-.05	.18
PE-QC 2	.20***	.35***	.28***	.21***	.25**	.12
NE-F	-.10*	-.04	-.10	.11	.07	.06
NE-H	.35***	.32***	.45***	.31***	.37***	.36***
NE-QC 1	.14**	.20**	-.00	-.05	-.11	.11
NE-QC 2	-.07	-.26***	-.15*	-.03	-.28**	-.05
F-H	-.24***	-.45***	-.09	-.19**	-.36***	-.35***
F-QC 1	.24***	.16**	.23***	.21***	.08	.03
F-QC 2	.06	-.08	.06	.22***	.09	.07
H-QC 1	.11*	-.01	.10	-.06	.07	.02
H-QC 2	-.18***	-.02	.03	-.11	.02	.04
QC1- QC2	.27***	.33***	.26***	.22***	.44***	.30***

*p < .05;

**p < .01;

***p < .001;

SM group = Spanish adolescents who evaluate Moroccan-origin youth; SR group = Spanish adolescents who evaluate Romanian-origin youth; SE group = Spanish adolescents who evaluate Ecuadorian-origin youth; M group = Moroccan-origin adolescents; R group = Romanian-origin adolescents; E group = Ecuadorian-origin adolescents; M = Morality; I = Immorality; S = Sociability; C = Competence; PE = Positive emotions; NE = Negative emotions; F = Facilitation; H = Harm; QC1 = Quantity of contact; QC2 = Quality of contact.

Network analysis

The centrality indices are shown in Table 4. Positive emotions appear as the most influential variable (with the highest values of positive expected influence indices; Hypothesis 2) in the positive direction of the network for Spanish adolescents who evaluated Romanian youth (1.65), Spanish adolescents who evaluated Ecuadorian youth (2.06), Moroccan (1.82), Romanian (1.37) and Ecuadorian (1.80) adolescents. This variable (positive emotions) was the second most influential for Spanish adolescents who evaluated Moroccan youth (1.12) behind sociability (1.74). The sociability dimension was the second most influential for Romanian (1.23) and Ecuadorian (1.13) adolescents, as was the quantity of contact for Spanish adolescents who evaluated Romanian youth (1.46).

Table 4

Centrality indices of each variable in the six sub-samples.

	SM group			SR group			SE group			M group			R group			E group
	EIn	Clos	Bet	EIn	Clos	Bet	EIn	Clos	Bet	EIn	Clos	Bet	EIn	Clos	Bet	EIn	Clos	Bet
M	-0.24	0.44	0.03	-0.14	-0.08	-0.34	-0.20	1.47	0.16	-0.42	0.40	0.31	-0.70	0.46	0.15	-0.80	0.67	0.83
I	-1.55	-0.76	-0.32	-1.22	0.55	1.03	-1.90	0.01	-0.48	-1.51	-0.64	-0.10	-1.80	0.11	-0.10	-1.27	-0.37	-0.09
S	1.74	1.07	1.09	0.59	0.49	1.03	0.37	0.86	-0.48	0.74	0.60	-0.92	1.23	1.10	0.91	1.13	1.15	0.52
C	-0.51	-1.67	-1.37	-0.03	-2.34	-1.38	0.15	-0.81	-1.12	0.09	-0.24	-0.92	0.18	-1.03	-1.37	-0.36	-1.35	-1.32
PE	1.12	1.42	0.39	1.65	1.30	1.72	2.06	0.99	1.75	1.82	1.71	1.54	1.37	1.50	1.67	1.80	1.45	1.44
NE	-0.36	-0.39	0.39	-0.20	0.75	0.34	-0.54	-0.12	1.43	-0.13	-0.63	-0.10	0.01	0.42	0.66	0.15	-0.72	-0.40
F	0.40	1.20	1.44	-0.71	-0.07	-0.34	0.41	0.29	0.16	0.10	1.36	1.74	-0.16	0.39	-0.61	0.20	1.07	1.13
H	-0.98	-0.18	0.74	-1.32	-0.59	-0.69	-0.63	-1.04	-0.80	-1.43	-0.36	0.31	-1.10	-0.71	-0.61	-1.33	-0.25	0.21
QC1	0.77	-0.61	-1.02	1.46	-0.44	-1.03	0.01	-1.81	-1.12	-0.04	-1.49	-0.92	0.21	-1.80	-1.37	-0.02	-0.97	-1.32
QC2	-0.37	-0.52	-1.37	-0.09	0.42	-0.34	0.27	0.16	0.48	0.77	-0.71	-0.92	0.74	-0.43	0.66	0.50	-0.69	-1.01

SM group = Spanish adolescents who evaluate Moroccan-origin youth; SR group = Spanish adolescents who evaluate Romanian-origin youth; SE group = Spanish adolescents who evaluate Ecuadorian-origin youth; M group = Moroccan-origin adolescents; R group = Romanian-origin adolescents; E group = Ecuadorian-origin adolescents; EIn = Expected influence; Clos = Closeness; Bet = Betweenness; M = Morality; I = Immorality; S = Sociability; C = Competence; PE = Positive emotions; NE = Negative emotions; F = Facilitation; H = Harm; QC1 = Quantity of contact; QC2 = Quality of contact; Numbers in bold indicate the highest scores for the different centrality indices.

In the negative direction, the immorality dimension appeared to be more influential (with the highest values of negative expected influence indices; Hypothesis 1) for Spanish adolescents who evaluated Moroccan youth (-1.55), Spanish adolescents who evaluated Ecuadorian youth (-1.90), and Moroccan (-1.51) and Romanian (-1.80) adolescents, followed by behavioral tendencies of harm for Moroccan (-1.43) and Romanian (-1.10) adolescents. Harm was the most influential variable for Spanish adolescents who evaluated Romanian youth (-1.32) and Ecuadorian adolescents (-1.33), followed by immorality for both groups (-1.22 and -1.27, respectively).

Regarding closeness indices, positive emotions appeared as the best-connected node within the network (with the highest values of closeness indices; Hypothesis 2) for Spanish adolescents who evaluated Moroccan youth (1.42), Spanish adolescents who evaluated Romanian youth (1.30), and Moroccan (1.71), Romanian (1.50) and Ecuadorian (1.45) adolescents. The positive emotions node was also the second best connected variable (0.99) behind morality (1.47) for Spanish adolescents who evaluated Ecuadorian youth. Scores in this index were followed by behavioral tendencies of facilitation for Spanish adolescents who evaluated Moroccan youth (1.20), and Moroccan adolescents (1.36), and by sociability for Romanian (1.10) and Ecuadorian (1.15) adolescents.

Finally, positive emotions appeared as the node with the highest scores in betweenness (there were more short paths that passed through this node than through the other ones; Hypothesis 2) for Spanish adolescents who evaluated Romanian and Ecuadorian youths (1.72 and 1.75, respectively) and for Romanian and Ecuadorian adolescents (1.67 and 1.44, respectively). This node was followed by sociability and immorality for Spanish adolescents who evaluated Romanian youth (1.03), by sociability for Romanian adolescents (0.91), by negative emotions for Spanish adolescents who evaluated Ecuadorian youth (1.43) and by behavioral tendencies of facilitation for Ecuadorian adolescents (1.13). For Spanish adolescents who evaluated Moroccan youth and for Moroccan-origin adolescents, facilitation behavior was the node with the higher scores in this index (1.44 and 1.74, respectively), followed by sociability for Spanish adolescents who evaluated Moroccan youth (1.09) and by positive emotions for Moroccan adolescents (1.54). The resulting network is displayed in Fig 1 (for Spanish adolescents) and Fig 2 (for adolescents with immigrant backgrounds). In general, for the six subsamples, we found a similar pattern in the structure of the network. The nodes corresponding to the stereotypes appear as a structure at the top of the networks in all samples. Similarly, nodes corresponding to positive and negative emotions appear below the stereotype structure. At the bottom of the networks, facilitation and harm nodes appear at both ends and are related to emotions. Finally, the contact quality node appears, also in the lower part of the network, and, in general, related to the facilitation and positive emotions nodes.

Fig 1

Estimated networks for Spanish adolescents who evaluate Moroccan (A), Romanian (B) and Ecuadorian (C) youths. M = Morality; I = Immorality; S = Sociability; C = Competence; PE = Positive emotions; NE = Negative emotions; F = Facilitation; H = Harm; QC1 = Quantity of contact; QC2 = Quality of contact.

Fig 2

Estimated networks for Moroccan (A), Romanian (B) and Ecuadorian (C) adolescents who evaluate Spanish youth. M = Morality; I = Immorality; S = Sociability; C = Competence; PE = Positive emotions; NE = Negative emotions; F = Facilitation; H = Harm; QC1 = Quantity of contact; QC2 = Quality of contact.

Morality, sociability and competence dimensions were directly and positively related to each other for Spanish adolescents who evaluated Moroccan, Romanian and Ecuadorian youths, and Moroccan and Romanian adolescents. For Ecuadorian adolescents, only morality and sociability were related. Sociability is the stereotype dimension that had the strongest direct relationship with positive emotions for Spanish adolescents who evaluated Moroccan and Romanian youths, and Moroccan and Romanian adolescents. For Spanish adolescents who evaluated Ecuadorian youth, morality was more strongly related to positive emotions than sociability. For Ecuadorian adolescents, only morality and competence were related to positive emotions. As can be seen on the network (Hypothesis 3), emotions act as a link between stereotype dimensions (at the top of the network) and behavioral tendencies (at the bottom and sides of the network): positive emotions act as a link between the positive stereotypical dimensions (sociability, morality and competence) and behavioral tendencies of facilitation whereas negative emotions act as a link between the immorality dimension and behavioral tendencies of harm.

Quantity and quality of contact were directly and positively related to each other in the six subsamples. Quality of contact was related to the rest of the network (Hypothesis 4), mainly through positive emotions for the Spanish adolescents who evaluated Moroccan and Ecuadorian youths, and Moroccan, Romanian and Ecuadorian adolescents. For the Spanish adolescents who evaluated Romanian youth, the quality of contact was related to the rest of the network through positive and negative emotions.

Discussion

The aim of this study was to analyze the relationship between different evaluative reactions of the intergroup attitudes (stereotypes, emotions, behavioral tendencies) and contact in Spanish adolescents evaluating different ethnic minorities (Moroccans, Romanians and Ecuadorians) and in immigrant-background adolescents evaluating Spanish youth, using empirical network analysis to explore the structure and dynamic properties of intergroup attitudes in these groups.

The centrality indices showed that immorality appears as the most influential stereotype dimension, following by sociability (above morality and competence). These results partially confirm our first hypothesis (H1). The result of the prominent role of immorality above the other dimensions (morality, sociability and competence) is innovative but not surprising [13-17]. Many previous studies have also shown a prominent role of morality in intergroup relations compared to other dimensions such as sociability and competence [10, 11, 53]. In our study, however, sociability had a more central role than morality. A possible explanation for the differences found with respect to previous studies could be the simultaneous measurement of both variables (morality and immorality) and the control of the effect of the rest of the evaluative reactions of attitudes due to the use of partial correlations in network analysis. In addition, most previous studies based on the SCM and its subsequent modifications have focused on the adult population [10]. In a study by Constantin and Cuadrado [23], which was conducted on an adolescent population, the results showed that sociability was moderately more important than morality in predicting positive emotions. It is possible that, for adolescents, sociability plays a more prominent role than morality, contrary to what happens with the adult population. Despite this, the present study is in line with previous findings on the distinction between sociability and morality and covered in the Moral Primacy Model [25].

In general, the results show similar structural patterns in the six studied groups: the positive emotions node was the most highly connected node (or central) with the rest of the network, and two types of emotions (positive and negative) acted as links between stereotypes and behavioral tendencies. These results are in line with the BIAS Map [5, 6], the IET [19] and research derived from these models [22] and confirming our second hypothesis (H2). The central role of emotions has also been found by Nariman et al. [30] from a network approach. In their study, they focused on stereotypical, emotional and behavioral evaluative responses toward Roma people (one of the main minority groups in Spain and in many European countries, associated with the stereotype of begging and delinquency, and characterized as a target group for discrimination, hate crimes and social exclusion [54-56]) and they observed that the central nodes are those of an affective nature (national identity, sympathy, and empathy).

In the present study, using network analysis, we found a relationship between stereotypes and behavioral tendencies through emotions, but this was compatible with the bidirectionality found in previous research [22] and defended by previous theories (IET [19]). They also did not find a direct relationship between sociability and positive emotions. In our study, on the contrary, the dimension of sociability was the closest and the one that presented the strongest direct relationship with positive emotions for the four groups in the study: Spanish adolescents evaluating Moroccan and Romanian youths, and Moroccan and Romanian origin adolescents evaluating Spanish youth. Even so, the direct relationship between morality and positive emotions did not appear in two of the studied groups (Spaniards evaluating Romanian youth and Romanian-origin adolescents) or was very week (for the rest of the groups).

Although morality and immorality dimensions are directly related, they also appear as dimensions that are part of different processes: while sociability (primarily) and morality and competence (to a lesser extent) are related to facilitation behaviors through positive emotions, immorality is related to harm behaviors through negative emotions, confirming our third hypothesis (H3). These results are in line with previous studies [26, 27], which pointed out the need to study positive and negative attitudes as separate dimensions. Including the stereotype dimension of immorality and analyzing its relationship with behaviors of harm through negative emotions is an important contribution of this study. Positive emotions also show a key role in the relationship between contact and intergroup attitudes. The results show some common patterns in the different groups. For the Spaniards evaluating Moroccan and Ecuadorian youths, and for the three groups of adolescents with an immigrant background (Moroccan, Romanian and Ecuadorian), the quality of contact was related to the rest of the network, mainly through positive emotions. For Spanish adolescents evaluating Romanian youth, this variable was related to the rest of the network through positive and negative emotions. These results partially confirm our hypothesis (H4) and they are in line with previous research [37, 38] showing that the quality of contact is related with different measures of intergroup attitude, especially affective reactions. The quality of contact seems to have a greater influence than the quantity (H4) for Spanish adolescents evaluating Ecuadorian youth and for the three groups of adolescents with an immigrant background. This result coincides with previous findings [37-39], showing that the quality of contact has a greater effect on intergroup attitudes than quantity. However, for the Spaniards who evaluated Moroccan youth, the quantity of contact had a greater centrality in the network than quality.

Taking into account the structure and dynamic properties of the networks and the estimated centrality indices, some conclusions can be drawn that may be of interest when planning interventions. Firstly, stereotypes relate to behavioral tendencies through two different pathways: through positive emotions for facilitation behaviors and through negative emotions for harm behaviors. To achieve successful intergroup relations involving cooperation and the development of friendly relationships (and not just a reduction of conflict), it would be appropriate to intervene in parallel in these two pathways. Secondly, due to the centrality of these variables and, therefore, their ability to affect the entire network, interventions should focus on positive emotions (and the perception of sociability) towards out-groups and decrease the perception of immorality. This could be an appropriate strategy to achieve positive overall attitudes and intergroup relationships with more and better contact. In this study, it is observed that intergroup contact quality plays a key role in improving intergroup attitudes. The direct relationship between the quality of contact and positive emotions (one of the most central variables of the network) indicates that achieving quality relationships could have an important effect on the whole attitudinal network. For future research, it would be of great interest to inquire about the people with whom adolescents have contact in class (whether they are from their own ethnocultural group or from other out-groups), including a social network analysis perspective [57, 58].

Surveying adolescents outside of the school context is very difficult, so the use of accessible samples (instead of a representative sample) and the sample size of some groups requires caution regarding the generalization of results to the rest of the adolescent population, which may be a limitation of the present study. In addition, it should be noted that all the measures used are self-reported, capturing only the self-perceptions of adolescents, with the drawbacks that this type of measure can have in capturing intergroup attitudes already described in the literature.

Despite these limitations, the present study contributes significantly to the psychosocial literature, using empirical network analysis, a recent and innovative perspective in the analysis of the structure of intergroup attitudes that allowed us to identify the key variables of this structure. Furthermore, this analysis was carried out with six groups of adolescents, from the majority (Spanish) and minority perspective (Moroccan, Romanian, and Ecuadorian origin). Our study shows similarities in the structure of intergroup attitudes and their key variables (e.g., the central role of emotions connecting stereotypes and behavioral tendencies), but also some differences depending on the group, thus supporting the specificity of intergroup attitudes depending on the context and related groups.

15 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Natasha McDonald, PhD

Associate Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. Thank you for stating the following in the Acknowledgments Section of your manuscript:

“This study is part of the project ‘Prejudiced attitudes, acculturation process and adjustment of immigrant and host adolescents’ [Reference PS2016-80123-P], funded by the Ministry of Economy, Industry and Competitiveness (Spain).”

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“This study is part of the project ‘Prejudiced attitudes, acculturation process and adjustment of immigrant and host adolescents’ [Reference PS2016-80123-P], funded by the Ministry of Economy, Industry and Competitiveness (Spain).

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors present an interesting article that draws on a large sample of Spanish adolescents (an understudied population given the topic) with necessary considerations taken regarding ethics and the surveying of minors. I truly enjoyed reading the article.

Here are some considerations that might help make it even better:

CONTEXT AND PARTICIPANTS

Some more on the context might be relevant. For non-Spanish readers “about 10% of the total Student body during the 2019-2020...” (p. 3) fails to highlight the 2000-2020 increase in the immigrant population at large and in schools within Spain. What are the numbers like in the specific area in Spain where data was collected? Relatedly, where were these public secondary schools located? In larger or smaller cities? Within what Autonomous Community or Autonomous Communities? Were all schools from the same city or county? Were they diverse in their percentage of immigrant population? How many total schools are we talking about? How is this area different or similar to other in Spain. Were the Spanish, Moroccan, Romanian and Ecuadorian groups surveyed similar the population of study? If so, could this be highlighted in a table for example? Also, certain language could use some clarification. For example, I understand what the authors mean by “Roma people” but some context for non-European readers might be helpful.

LITERATURE REVIEW

Overall the citations seem a bit dated and few studies in Spain are cited. There is some high-quality recent (2018-2021) research coming out of nearby Catalonia (Spain) that I think the authors might want to read and integrate. These studies also cite other relevant research you might want to look into, for example:

Bobowik, M., Benet-Martínez, V., & Repke, L. (2021). “United in diversity”: The interplay of social network characteristics and personality in predicting outgroup attitudes. Group Processes & Intergroup Relations, https://doi.org/10.1177/13684302211002918

Bobowik, M., Benet‐Martínez, V., & Repke, L. (2021, online first). Ethnocultural diversity of immigrants' personal social networks, bicultural identity integration and global identification. International Journal of Psychology. https://doi.org/10.1002/ijop.12814

Repke, L., & Benet-Martínez, V. (2018). The (diverse) company you keep: Content and structure of immigrants’ social networks as a window into intercultural relations in Catalonia. Journal of Cross-Cultural Psychology, 49(6), 924-944.

Wilson-Daily, A. E., & Kemmelmeier, M. (2021). Who is on our side? complexities of national identification among native and immigrant youth in Catalonia. Journal of Youth Studies, 24(8), 994-1014.

Wilson-Daily, A. E., Kemmelmeier, M., & Prats, J. (2018). Intergroup contact versus conflict in Catalan high schools: A multilevel analysis of adolescent attitudes toward immigration and diversity. International Journal of Intercultural Relations, 64, 12-28.

Also, a bit dated is the reference to Allport. Athough Pettigrew and Tropp (2006) is mentioned, their building off of and improvement of Allport’s original work is not referenced. Pettigrew and Tropp expanded Allport’s original hypothesis and highlighted necessities for its accomplishment. Nevertheless, this is not highlighted in any way in the article. Furthermore, the authors might be interested in consulting and considering some research from Hewstone and colleagues, such as:

Laurence, J., Schmid, K., Rae, J. R., Hewstone, M. (2019). Prejudice, contact, and threat at the diversity–segregation nexus: A cross-sectional and longitudinal analysis of how ethnic out-group size and segregation interrelate for inter-group relations. Social Forces, 97, 1029–1066. https://doi.org/10.1093/sf/soy079

Wölfer, R., Faber, N. S., Hewstone, M. (2015). Social network analysis in the science of groups: Cross-sectional and longitudinal applications for studying intra- and intergroup behavior. Group Dynamics: Theory, Research, and Practice, 19, 45–61. https://doi.org/10.1037/gdn0000021

Wölfer, R., Hewstone, M. (2017). Beyond the dyadic perspective: 10 reasons for using social network analysis in intergroup contact research. British Journal of Social Psychology, 56, 609–617. https://doi.org/10.1111/bjso.12195

These should be at least read (if not cited) and perhaps contemplated in regard to study limitations. I see that you do reference a 1993 article of Hewstone’s but I would encourage a more in-depth reading of his group’s more recent work.

ANALYSES

I do not see any control for example for socioeconomic status, given the importance that this plays on intergroup attitudes, the autors might want to comment on this. If this data is available I would find a way to include it in the analyses.

Also, if students are nested in classrooms, which are then nested in schools, and then neighborhoods, a study on intergroup contact might want to contemplate multilevel modeling. Was this data collected (i.e., who is in what classroom/School)? The analysis might benefit from “who am I in contact with in this classroom /school” in ADDITION to the self-reported “Quality of contact”.

OTHER, MORE MINOR QUESTIONS OR CONSIDERATIONS

If students did not identify as Spanish, Moroccan, Romanian or Ecuadorian did they not fill out a questionnaire? How were the diferent survey options presented to the students?

Reviewer #2: The manuscript investigates the perception of different ethnic minorities in light of the SCM and Bias Map. The manuscript is well written and easy to follow. The study considers un understudied sample (adolescents) and employed an innovative approach (network analysis). Therefore, I am happy to recommend this paper for publication after a revision.

1) The distinction between sociability and morality has been dismissed lightly. I would suggest the authors better describe the innovative value of their theoretical approach. They could describe the moral primacy model of impression development (Brambilla et al., 2021; Goodwin, 2015) and state how the present study fits/complements/extends prior insights on the distinction between sociability and morality.

Brambilla, M., Sacchi, S., Rusconi, P., Goodwin, G. (2021). The primacy of morality in impression development: Theory, research, and future directions. Advances in Experimental Social Psychology, 64, 187-262.

Goodwin, G. P. (2015). Moral character in person perception. Current Directions in Psychological Science, 24, 38-44.

2) Hypotheses: The distinction between morality and immorality is well explained in the general discussion. However, I would suggest describing such a distinction only in the introduction. This would be functional to better understand the hypotheses. I would refer to prior work on the confirmability of traits-concepts (Reeder & Brewer, 1979).

Reeder, G. D., & Brewer, M. B. (1979). A schematic model of dispositional attribution in interpersonal perception. Psychological Review, 86, 61.

3) I do not immediately understand why the authors measured the negative and the positive pole of morality, while sociability and competence were measured by considering only the positive pole. Please explain.

4) I would also suggest describing more in detail how morality has been conceptualized.

Morality is a broad construct defined by distinct components (see for instance the distinction between deontology and consequentialism - Sacchi et al., 2014; or the distinction between harm and fairness, Gray et al., 2009).

Sacchi, S., Riva, P., Brambilla, M., & Grasso, M. (2014). Moral reasoning and climate change mitigation: The deontological reaction toward the market-based approach. Journal of Environmental Psychology, 38, 252-261.

Gray, K., & Wegner, D. M. (2009). Moral typecasting: divergent perceptions of moral agents and moral patients. Journal of personality and social psychology, 96(3), 505.

Reviewer #3: This is a very interesting paper that contributes to the intergroup attitudes literature. It is well-performed and results are clearly exposed.

As for suggestions that could improve the paper are: including in the abstract the application of the results, deepen on the conclusions as for the implications of the results, and performing further analyses to guarantee the comparability of first and secon-generation of the group members. Moreover, were there any difference by different schools? maybe the incomes, sex or tother variables should also be considered.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes: Esther Lopez-Zafra

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Plos One Review.docx

Click here for additional data file.

28 Apr 2022

Dear Editor and Reviewers,

Thank you very much for the consideration and the feedback you have provided to our manuscript. We have introduced your recommendations in this new version and we explain how we addressed each advice or question in the following pages.

REVIEWER #1:

The authors present an interesting article that draws on a large sample of Spanish adolescents (an understudied population given the topic) with necessary considerations taken regarding ethics and the surveying of minors. I truly enjoyed reading the article. Here are some considerations that might help make it even better:

Context and participants

Some more on the context might be relevant. For non-Spanish readers “about 10% of the total Student body during the 2019-2020...” (p. 3) fails to highlight the 2000-2020 increase in the immigrant population at large and in schools within Spain. What are the numbers like in the specific area in Spain where data was collected? Relatedly, where were these public secondary schools located? In larger or smaller cities? Within what Autonomous Community or Autonomous Communities? Were all schools from the same city or county? Were they diverse in their percentage of immigrant population? How many total schools are we talking about? How is this area different or similar to other in Spain. Were the Spanish, Moroccan, Romanian and Ecuadorian groups surveyed similar the population of study? If so, could this be highlighted in a table for example?

We have added some information in the manuscript as detailed in the following paragraphs about a more in-depth description of the context in which the study was conducted. The questionnaires were applied in 17 secondary schools located in five provinces in Spain (Almería, Alicante, Castellón, Madrid and Murcia), selected because they are among those with the highest percentages of foreign population from any of the three origins (Moroccan, Romanian or Ecuadorian). Specifically, the percentages of foreign population in these provinces were 21.37%, 19.84%, 15.18%, 14.15% and 14.64%, respectively (National Institute of Statistics, [47]).

All the municipalities in which the questionnaires were applied has less than 100000 inhabitants, except Puente de Vallecas (Madrid) with 240000 inhabitants. Official data on the percentages of foreign adolescents enrolled in all secondary schools was not available.

Information included in the manuscript:

Introduction section (lines 26-30): “The increase in diversity in current societies is reflected in the presence of students with an immigrant background in secondary schools (in Spain, about 10% of the total student body during the 2019-2020 academic year whereas this percentage was less than 2% in the 2000-2001 academic year, according to the Spanish Ministry of Education, Culture and Sport [1, 2]).”

Participants section (lines 201-202): “They were enrolled in 17 public secondary schools in different municipalities in five Spanish provinces (Almería, Alicante, Castellón, Madrid and Murcia).”

Procedure section (lines 265-269): “After carrying out a residential census analysis, five sampling zones (Almería, Alicante, Castellón, Madrid and Murcia) were chosen with the largest immigrant population [34]. Specifically, the percentages of foreign population in these provinces were 21.37%, 19.84%, 15.18%, 14.15% and 14.64%, respectively according to the National Institute of Statistics [48].”

Also, certain language could use some clarification. For example, I understand what the authors mean by “Roma people” but some context for non-European readers might be helpful.

A clarification has been included in parentheses (lines 412-414): “…Roma people (one of the main minority groups in Spain and in many European countries, associated with the stereotype of begging and delinquency, and characterized as a target group for discrimination, hate crimes and social exclusion [54-56])…”

Literature review

Overall the citations seem a bit dated and few studies in Spain are cited. There is some high-quality recent (2018-2021) research coming out of nearby Catalonia (Spain) that I think the authors might want to read and integrate. These studies also cite other relevant research you might want to look into, for example:

Bobowik, M., Benet-Martínez, V., & Repke, L. (2021). “United in diversity”: The interplay of social network characteristics and personality in predicting outgroup attitudes. Group Processes & Intergroup Relations, https://doi.org/10.1177/13684302211002918

Bobowik, M., Benet‐Martínez, V., & Repke, L. (2021, online first). Ethnocultural diversity of immigrants' personal social networks, bicultural identity integration and global identification. International Journal of Psychology. https://doi.org/10.1002/ijop.12814

Repke, L., & Benet-Martínez, V. (2018). The (diverse) company you keep: Content and structure of immigrants’ social networks as a window into intercultural relations in Catalonia. Journal of Cross-Cultural Psychology, 49(6), 924-944.

Wilson-Daily, A. E., & Kemmelmeier, M. (2021). Who is on our side? complexities of national identification among native and immigrant youth in Catalonia. Journal of Youth Studies, 24(8), 994-1014.

Wilson-Daily, A. E., Kemmelmeier, M., & Prats, J. (2018). Intergroup contact versus conflict in Catalan high schools: A multilevel analysis of adolescent attitudes toward immigration and diversity. International Journal of Intercultural Relations, 64, 12-28.

Thank you very much for the recommendation of these interesting papers. We have integrated the results of some of these studies in our manuscript:

Lines 155-164: “Bobowik et al. [40] conducted a study using a social network approach in Spain with adults of immigrant background and found that the proportion of intergroup contacts (with native-born population) representing strong ties (i.e., close relationships such as friends, romantic partners, and immediate and extended family members) and ethnic diversity among these ties were associated with more favorable attitudes toward outgroups. In another study conducted in the same community with 15- and 16-year-old students, it was observed that students who reported frequent interaction with other students of different nationalities and religions held more positive intergroup attitudes and the presence of a greater number of immigrants in classrooms was related to lower levels of xenophobia [41].”

Also, a bit dated is the reference to Allport. Athough Pettigrew and Tropp (2006) is mentioned, their building off of and improvement of Allport’s original work is not referenced. Pettigrew and Tropp expanded Allport’s original hypothesis and highlighted necessities for its accomplishment. Nevertheless, this is not highlighted in any way in the article. Furthermore, the authors might be interested in consulting and considering some research from Hewstone and colleagues, such as:

Laurence, J., Schmid, K., Rae, J. R., Hewstone, M. (2019). Prejudice, contact, and threat at the diversity–segregation nexus: A cross-sectional and longitudinal analysis of how ethnic out-group size and segregation interrelate for inter-group relations. Social Forces, 97, 1029–1066. https://doi.org/10.1093/sf/soy079

Wölfer, R., Faber, N. S., Hewstone, M. (2015). Social network analysis in the science of groups: Cross-sectional and longitudinal applications for studying intra- and intergroup behavior. Group Dynamics: Theory, Research, and Practice, 19, 45–61. https://doi.org/10.1037/gdn0000021

Wölfer, R., Hewstone, M. (2017). Beyond the dyadic perspective: 10 reasons for using social network analysis in intergroup contact research. British Journal of Social Psychology, 56, 609–617. https://doi.org/10.1111/bjso.12195

These should be at least read (if not cited) and perhaps contemplated in regard to study limitations. I see that you do reference a 1993 article of Hewstone’s but I would encourage a more in-depth reading of his group’s more recent work.

Information regarding intergroup contact has been extended in the manuscript (lines 137-143): “equal status between the groups, common goals, intergroup cooperation, and the support of authorities, law, or custom). However, in Pettigrew and Tropp’s meta-analysis, it was observed that these conditions were not necessary for a reduction in prejudice to occur and these authors proposed that the simple familiarity that occurs with contact generates liking. In addition, they point out that the relationship between familiarity and liking could be influenced by variables such as the reduction of uncertainty, intergroup anxiety and threat”.

Thank you very much for the recommendations of these articles. We have included one of them in the Introduction section (lines 143-147): “The study by Laurence et al. [36] showed that increased out-group size was related to intergroup attitudes towards them only in more segregated communities (but not in the case of more integrated communities), and that this is explained by an increase in perceived threat and a decrease in the quality of contact (when the increase in out-group size exceeds a certain value).”

The other two papers have been mentioned in future lines of research (lines 464-467): “For future research, it would be of great interest to inquire about the people with whom adolescents have contact in class (whether they are from their own ethnocultural group or from other out-groups), including a social network analysis perspective [57, 58].”

We have not described these articles in greater depth as they focus on social network analysis, which encompasses network analysis from a psycho-sociological point of view (of great interest to social psychology), while the empirical network analysis we have used is approached from a more psychometric focus with variables of great impact in the psychosocial literature. However, due to their interest, we consider it important to consider them as future lines of research.

Analyses

I do not see any control for example for socioeconomic status, given the importance that this plays on intergroup attitudes, the authors might want to comment on this. If this data is available I would find a way to include it in the analyses.

We have a measure of perceived socioeconomic status in the study (the scores of this variable range from 1 to 10). Following the suggestion of the reviewer, we did perform the corresponding analyses to explore if there were differences between the groups in this variable. The average score is 6.38 (SD = 1.36) for SM group, 6.39 (SD = 1.29) for SR group, 6.43 (SD = 1.33) for SE group, 6.14 (SD = 1.61) for M group, 6.36 (SD = 1.28) for R group, and 6.69 (SD = 1.36) for E group. Results showed that differences between groups appeared only between the E group (Ecuadorian adolescents) and the rest of the groups (p < .001 with SM, SR, SE groups and p < .01 with M and R groups).

Results of the ANOVA (See attached file titled Response to Reviewers)

One of the ways to include a variable as a covariate in network analysis would be to include this variable as another node within the network and see if substantial changes occur in the proposed network. We have included the perceived socioeconomic status variable in our network analysis (See attached file titled Response to Reviewers)

Perceived socioeconomic status appears, in general, unlinked to the rest of the network and does not produce important changes in the structure of attitudes. There is only a stronger relationship between perceived socioeconomic status and quantity of contact in the SR group (the higher the perceived status of Spanish adolescents, the lesser contact they have with adolescents of Romanian origin). However, the network continues to maintain the same structure presented without introducing it.

For these reasons, we have decided not to include this information in the manuscript so as not to extend the length of the manuscript excessively. However, if the reviewer considers it appropriate, we could add it in a footnote.

In the participants' section of the new version of the article we have included descriptive data (means and standard deviations) on perceived socioeconomic status, where it can be seen that the scores are similar in all the groups studied (lines 208-211): “Average perceived socioeconomic status score (ranging from 1 to 10 points) was 6.38 (SD = 1.36) for SM group, 6.39 (SD = 1.29) for SR group, 6.43 (SD = 1.33) for SE group, 6.14 (SD = 1.61) for M group, 6.36 (SD = 1.28) for R group, and 6.69 (SD = 1.36) for E group”.

Also, if students are nested in classrooms, which are then nested in schools, and then neighborhoods, a study on intergroup contact might want to contemplate multilevel modeling. Was this data collected (i.e., who is in what classroom/School)? The analysis might benefit from “who am I in contact with in this classroom /school” in ADDITION to the self-reported “Quality of contact”.

We have the school and grade level data for each student (not the exact classroom). However, one of the aims of this study was the implementation of psychological network analysis. We know and agree with the reviewer in regards to the usefulness of using multilevel analyses, but using these analyses would imply approaching this study from a completely different methodological perspective than the proposed objective. We agree with the reviewer that "the analysis might benefit from who am I in contact with in this classroom /school, in addition to the self-reported quality of contact”. But the research design did not include the collection of these data (only quality of contact). However, we have included a comment on this in the discussion section, pointing out limitations and future lines of research.

Other, more minor questions or considerations

If students did not identify as Spanish, Moroccan, Romanian or Ecuadorian did they not fill out a questionnaire? How were the diferent survey options presented to the students?

All students who were in class and had parental permission filled out the questionnaire. Before the questionnaire was handed out, the adolescents answered two questions (in paper and pencil format) about their origin (country of birth and country of birth of their parents, both mother and father). In this way, the researchers determined their origin. If they were Spanish adolescents (from Spanish parents) or from any of the origins included in this study (Moroccan, Rumanian or Ecuadorian), they were administered the corresponding questionnaire. If they were from any other ethnocultural background, we administered a different questionnaire that asked in a general way about the country of origin of their parents (taking into account that many of them could be second generation). An example of a question in this general questionnaire was: “To what degree do you currently maintain the customs of your parents’ country of origin?” The data from the participants who responded to this general questionnaire have not been included in the present study. We have not included this information in the manuscript so as not to extend the length of the manuscript excessively. However, if the reviewer considers it appropriate, we could briefly add it in a footnote in Procedure section.

REVIEWER #2:

The manuscript investigates the perception of different ethnic minorities in light of the SCM and Bias Map. The manuscript is well written and easy to follow. The study considers un understudied sample (adolescents) and employed an innovative approach (network analysis). Therefore, I am happy to recommend this paper for publication after a revision.

1) The distinction between sociability and morality has been dismissed lightly. I would suggest the authors better describe the innovative value of their theoretical approach. They could describe the moral primacy model of impression development (Brambilla et al., 2021; Goodwin, 2015) and state how the present study fits/complements/extends prior insights on the distinction between sociability and morality.

Brambilla, M., Sacchi, S., Rusconi, P., Goodwin, G. (2021). The primacy of morality in impression development: Theory, research, and future directions. Advances in Experimental Social Psychology, 64, 187-262.

Goodwin, G. P. (2015). Moral character in person perception. Current Directions in Psychological Science, 24, 38-44.

Some information on the Moral Primacy Model has been added in the version of the manuscript (lines 86-92): “Some of the above findings have been captured in the three key assumptions of the Moral Primacy Model [25]: 1) Morality, sociability, and competence are distinct dimensions that make unique contributions to impression formation; 2) Morality has a primary role in the formation of impressions and the evaluations we make about other people and groups both at different stages of impression formation and in behavioral outcomes; 3) The primary role of morality is due to its close link to the judgment of whether other social targets represent an opportunity or a threat.”

The following information has also been included in the Discussion section (lines 402-404): “Despite this, the present study is in line with previous findings on the distinction between sociability and morality and covered in the Moral Primacy Model [25].”

2) Hypotheses: The distinction between morality and immorality is well explained in the general discussion. However, I would suggest describing such a distinction only in the introduction. This would be functional to better understand the hypotheses. I would refer to prior work on the confirmability of traits-concepts (Reeder & Brewer, 1979).

Reeder, G. D., & Brewer, M. B. (1979). A schematic model of dispositional attribution in interpersonal perception. Psychological Review, 86, 61.

Part of the information that appeared in the Discussion section has been moved to the Introduction section and the reference to the article recommended has been included (lines 58-61): “Previous research [13-17] has shown evidence of the theoretical importance and the diagnostic capacity of the negative pole of the morality domain (i.e., immorality) in the out-group’s impression formation (i.e., for this dimension, negative information is more enlightening than positive information).”

3) I do not immediately understand why the authors measured the negative and the positive pole of morality, while sociability and competence were measured by considering only the positive pole. Please explain.

Some studies (e.g., Goodwin & Darley, 2012; Sayans-Jiménez et al., 2017) have shown the importance of also considering negative aspects of morality (i.e., immorality) in impression formation. This has not currently been shown with the sociability and competence dimensions. Moreover, the study by Sayans-Jiménez et al. (2017) showed that adding negative items to the sociability and competence dimensions worsened the psychometric properties of a scale for measuring stereotype content. The decision to consider morality and immorality dimensions as separate dimensions is supported by previous studies such as Pittinsky et al. (2011) or Cacioppo & Berntson (1994) and those mentioned in the manuscript.

4) I would also suggest describing more in detail how morality has been conceptualized. Morality is a broad construct defined by distinct components (see for instance the distinction between deontology and consequentialism - Sacchi et al., 2014; or the distinction between harm and fairness, Gray et al., 2009).

Sacchi, S., Riva, P., Brambilla, M., & Grasso, M. (2014). Moral reasoning and climate change mitigation: The deontological reaction toward the market-based approach. Journal of Environmental Psychology, 38, 252-261.

Gray, K., & Wegner, D. M. (2009). Moral typecasting: divergent perceptions of moral agents and moral patients. Journal of personality and social psychology, 96(3), 505.

Thank you very much for the recommendation of these interesting papers. Because the conceptualization of morality is so well established in the psychosocial literature on the stereotype content and models collected in our study (and so as not to overextend the length of the paper) we have decided not to add this information to the manuscript.

We sincerely thank the reviewers for their valuable comments and recommendations, which have undoubtedly contributed to improving and clarifying our article.

REVIEWER #3:

This is a very interesting paper that contributes to the intergroup attitudes literature. It is well-performed and results are clearly exposed. As for suggestions that could improve the paper are:

Including in the abstract the application of the results

The following information has been added to the abstract (lines 32-37): “This could indicate that, to achieve successful intergroup relations involving cooperation and the development of friendly relationships, it would be appropriate to intervene in parallel in these two pathways. Due to the centrality of positive emotions (and sociability and immorality) and, therefore, their capacity to affect the entire network, focusing interventions on these variables could be an appropriate strategy to achieve overall positive attitudes.”

Deepen on the conclusions as for the implications of the results

Thank you very much for the suggestion. The implications of the study have been extended to the role of the quality of intergroup contact (lines 461-467): “In this study, it is observed that intergroup contact quality plays a key role in improving intergroup attitudes. The direct relationship between the quality of contact and positive emotions (one of the most central variables of the network) indicates that achieving quality relationships could have an important effect on the whole attitudinal network.”

Performing further analyses to guarantee the comparability of first and second-generation of the group members.

We did not have an ample enough sample to divide adolescents of immigrant origin into first and second generation. However, following the reviewer's suggestion, we carried out t-tests, comparing the mean scores obtained by both generations in all the study variables. The results of these analyses can be seen below. We found that there were no statistically significant differences in practically any of the variables included in the study, so we decided to use both generations together.

The results of the t-tests for each of the groups can be seen in the attached file titled Response to Reviewers.

Moreover, were there any difference by different schools? maybe the incomes, sex or other variables should also be considered.

Thank you very much for the suggestion. Our objective was to compare these empirical network models in different groups of adolescents as a function of their ethnocultural group (or the ethnocultural group that they evaluate). Therefore, our data collection was aimed at this objective and we do not have an ample enough sample to compare based on the variables that you propose. Despite this, and regarding the specific variables you mention.

We do not have a measure of income, but we do have a measure of perceived socioeconomic status. It can be seen from the responses to reviewer #1 that there are hardly any differences in perceived status between the different groups and that this variable does not produce important changes in the networks.

Regarding sex, and despite the fact that, as we have mentioned, we do not have an ample enough sample to make these comparisons nor was it our objective, we explain that the percentage of boys and girls is more or less balanced in all groups (around 50%).

Thanks again for your suggestions. We believe that they contribute substantially to improve the manuscript.

Best regards

The authors

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

1 Jun 2022

20 Jun 2022

Dear Editor and Reviewers,

Thank you very much for your comments. Your recommendations have been included in the new version. The changes are explained below.

Reviewer 1 felt that the manuscript has improved greatly, however, s/he is not able to assess whether the data analysis is conducted to the satisfactory level. I also am not sure how your results address the hypotheses. Please assist the readers in revising the part of the results. That is, how are the current results respond to the hypotheses listed on p. 9. Please refrain from using abbreviations, it makes readers very difficult to follow the results.

We have made some changes in the results section by linking each result to the corresponding hypothesis and eliminating all abbreviations.

In addition, the following clarification has been added regarding the structure of the networks (lines 353-358):

The nodes corresponding to the stereotypes appear as a structure at the top of the networks in all samples. Similarly, nodes corresponding to positive and negative emotions appear below the stereotype structure. At the bottom of the networks, facilitation and harm nodes appear at both ends and are related to emotions. Finally, the contact quality node appears, also in the lower part of the network, and, in general, related to the facilitation and positive emotions nodes.

Abbreviations have also been removed in the discussion section.

We hope that this will make the reading of the results clearer and simpler (all these changes have been introduced with change control in the manuscript).

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

We have checked the list of references to ensure that it is complete and correct. We have no citations for papers that have been retracted.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

30 Jun 2022

Intergroup attitudes and contact between Spanish and immigrant-background adolescents using network analysis.

PONE-D-21-14451R2

Dear Dr. Sánchez Castelló,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

I-Ching Lee

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

14 Jul 2022

PONE-D-21-14451R2

Intergroup attitudes and contact between Spanish and immigrant-background adolescents using network analysis

Dear Dr. Sánchez-Castelló:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. I-Ching Lee

Academic Editor

PLOS ONE