Measuring Bullying and Victimization Among Immigrant and Native Primary School Students: Evidence From Italy

Abstract

Italy is a particularly interesting context in which to study the phenomenon of bullying given the steadily increasing number of immigrant students attending Italian primary schools. We examined the psychometric properties of a short self-report measure of bullying and victimization across groups of students with various migration backgrounds. We then estimated, by latent mean comparisons, the rates of prevalence of bullying and victimization among different generations of immigrants and native students. Results concerning the factor structure of the measure were consistent with studies in other cultural contexts and complete scalar measurement invariance was found across immigrant backgrounds. The analyses showed that both first- and second-generation immigrant pupils reported being victimized more frequently than their native peers. However, the incidence of victimization for second generations was lower than that for first generations. Finally, no differences across different generations of immigrants and native students were found in reported bullying behaviors.

Keywords

children bullying victimization immigrant students Italy

Introduction

According to a widespread definition of bullying, students are bullied or victimized at school if one or more persons subjects them, repeatedly and over time, to negative actions with a harmful intention (by means of verbal or physical abuse, or in various other ways), and it is difficult for victims to defend themselves against these actions (Olweus, 1994). Bullying and victimization are serious problems in multicultural contexts (Juvonen & Graham, 2014). For immigrant students in particular it is crucial to establish positive relationships at school, in order for them to be well integrated into a new society (Berry, 2001; Berry, Phinney, Sam, & Vedder, 2006; Berry, Poortinga, Segall, & Dasen, 2002; Fandrem, Ertesvåg, & Strohmeier, 2010; Fandrem, Strohmeier, & Roland, 2009). Italy is particularly interesting for the study of such phenomena, as there has been a sharp increase in the number of foreign students over the last 15 years in this country. In fact, in the school year 2001-2002, there were 196,414 foreign students in Italian schools of all levels, corresponding to 2.2% of the total student population, whereas in the 2013-2014 school year the number of pupils without Italian citizenship had risen to 802,844, amounting to 9% of the total (Foundation for Initiatives and Studies on Multi-Ethnicity [ISMU], 2015). In 2013-2014, the number of foreign students in Italian primary schools was 283,233, amounting to 10% of the total number of primary school students (ISMU, 2015).

The immigrant population in Italy is characterized by several risk factors that are known from the literature to be associated with involvement in bullying and victimization. A first factor is related to academic achievement (e.g., Woods & Wolke, 2004): Immigrant students in Italy tend to perform less well at school (e.g., Alivernini, Manganelli, & Lucidi, 2016), and to have lower levels of civic knowledge (Alivernini & Manganelli, 2011; National Institute for the Evaluation of the Education System [INVALSI], 2011). A second risk factor is related to deviation from classroom norms (Juvonen & Graham, 2014). In fact, immigrant students in Italy tend to be more socially isolated in class (Alivernini & Manganelli, 2016), and they appear to be more oriented toward competition at school (Alivernini et al., 2018) than their peers. Finally, there is some evidence that Italian students have mixed attitudes toward immigrants (Cavicchiolo, Alivernini, & Manganelli, 2016), and this might affect their relationships with pupils coming from different cultural backgrounds. All these data suggest that in Italy, the problem of bullying and victimization in multicultural contexts is likely to be particularly relevant.

Bullying and Victimization Across Immigrants and Native Students

International research comparing the amount of bullying and being bullied in cultural contexts involving both native and immigrant students has had some mixed results (Boulton, 1995; Eslea & Mukhtar, 2000; Fandrem et al., 2010; Moran, Smith, Thompson, & Whitney, 1993; Strohmeier & Spiel, 2003). A number of studies showed no differences between natives and immigrants (Boulton, 1995; Eslea & Mukhtar, 2000; McKenney, Pepler, Craig, & Connolly, 2006; Moran et al., 1993; Verkuyten & Thijs, 2002, 2006). Various studies (Graham & Juvonen, 2002; Strohmeier & Spiel, 2003) found a higher incidence of bullying among natives than among immigrants, while Fandrem and colleagues (2009) found that the opposite was the case. Finally, some studies (Fu, Land, & Lamb, 2012; Graham & Juvonen, 2002; Strohmeier, Kärnä, & Salmivalli, 2011) showed that immigrants have a higher risk of being victimized. In the Italian context, research on bullying has focused mainly on middle and high school students (e.g., Baldry, 1998; Genta, Menesini, Fonzi, Costabile, & Smith, 1996; Vieno et al., 2015), while its incidence in immigrant and native groups of students in primary school has not yet been compared.

Research on bullying in Italy has shown that this phenomenon might be particularly relevant in this country (Genta et al., 1996). A recent study by Vieno and colleagues (2015) focused on middle and secondary school students and provided evidence for a decrease in students’ involvement in bullying behavior from 2002 to 2010. However, at the moment, there is no specific information regarding bullying among immigrant and Italian students in primary school.

In trying to correctly gauge the phenomenon of bullying among students with different migration backgrounds, a number of issues have to be taken into consideration. First of all, estimates need to be based on large samples that should be as representative as possible, given the fact that results based on small groups of students might well be very far from the population values as a whole. Second, the differences between first- and second-generation students should be analyzed, as one might suppose that second-generation students born in the host country are more likely to be socially integrated, and differences between them and native students might be attenuated (Barban & White, 2011; Neidert & Farley, 1985). This in turn could affect estimates of the relative incidence of bullying. The possible bias caused by the variance of measures across groups of students with different migration backgrounds is a further issue. This point is especially important because estimates of bullying and victimization in large samples are usually based on self-report measures, and the items involved might not be interpreted in the same way across groups with different countries of origin (Milfont & Fischer, 2015).

The Present Study

In the present study, we have taken the above-mentioned issues into account and, using the widespread self-report measure of bullying and victimization developed by Roland and Idsøe (2001), we have tried to make a contribution to the literature on the basis of data gathered from a national representative sample of Italian fifth graders. The aim of the study was twofold: We have analyzed the psychometric properties of the self-report measure to provide a perspective on bullying in the context of a country with a high and increasing presence of immigrant students at school. Second, on the basis of the comparisons of latent means, we have estimated the rates of prevalence of bullying and victimization among different generations of immigrants and native students to provide a reliable description of these phenomena in primary school. In the present study, the following research questions were addressed:

Research Question 1: What are the psychometric properties of a widespread self-report measure of bullying and victimization in the Italian cultural context?

Research Question 2: What is the level of invariance of the adopted self-report measure for studying bullying and victimization across students’ immigrant backgrounds?

Research Question 3: Are there differences in bullying between first-generation, second-generation, and native students?

Research Question 4: Are there differences in victimization between first-generation, second-generation, and native students?

Alongside these general research questions, the study also empirically evaluated some more specific hypotheses. The first one is based on the evidence of higher levels of social isolation of immigrant children at school compared with their native peers (Alivernini & Manganelli, 2016). On the basis of this fact, we would expect immigrant students to be more at risk of victimization than Italian students. The second hypothesis is that second-generation students should be at less risk of victimization than first-generation students because they are more likely to be socially integrated (Barban & White, 2011; Neidert & Farley, 1985). Given the mixed results reported in the literature, no specific hypothesis was formulated about bullying behaviors.

Method

Sample and Procedure

The data analyzed in the present study came from a sample of 26,670 Italian fifth-grade students who took part in the National Evaluation of Learning in May 2014 (INVALSI, 2014). Participating students were sampled from the whole population of Italian primary schools, and they were a nationally representative sample of fifth-grade students. Full classes of fifth graders were randomly selected and all the students in each of these classes were assessed using the survey. Each school dealt with the process of informed consent and parental permission according to the National Evaluation of Learning assessment protocol (INVALSI, 2014). Students were also given a standardized introduction, which informed them of the purpose of the study and gave instructions on how to complete the questionnaire. All the participating students completed the anonymous questionnaires in class during the first part of an ordinary school day.

The average age of the students was 10.36 years (SD = 0.56), 49% of the students were female, 6.5% were second-generation immigrants, and 3.3% were first-generation immigrants. In the present study, immigration status was defined in line with the criteria adopted by the Organisation for Economic Co-Operation and Development (2014). Students were thus considered second-generation immigrants if they were born in Italy, but with parents who were born in another country, while they were defined as first-generation immigrants if they were foreign-born and with foreign-born parents. Immigrant pupils in Italy mostly come from other European countries, and in the year of reference of the present study, Romanian and Albanian students alone represented one third of all foreign students in primary school (ISMU, 2015).

Instrument

The measures used in the present study were adapted from the scales developed by Roland and Idsøe (2001). In the instrument, the students were first given a standard written explanation of the basic components of bullying behavior, such as harming the victim by intention, repetition of the behavior over time, and the imbalance of power between the victim and the bully. The students were provided with the definition of bullying formulated by Midthassel, Bru, and Idsoe (2008):

We call it bullying when one or more students are unfriendly or unpleasant to a pupil that cannot defend him-/herself easily. This could include kicking, hitting, or shoving the pupil. It is also bullying when pupils are teased or when pupils are shut out from the others (p. 95).

The scales concerning “Bullying” and “Victimization” each consisted of four items addressing general and specific bullying behaviors (Table 1). The possible answers were as follows: 0 (never), 1 (now and then), 2 (weekly), and 3 (daily). The two scales were translated and then back-translated for use in the Italian sample.

Table 1.

The Items of the Bullying and Victimization Scales.

Bullying scale

This school year how often have you . . .

1. Bullied/hassled other students at school?

2. Bullied/hassled other students at school by teasing them?

3. Bullied/hassled other pupils at school by isolating them or shutting them out from others?

4. Bullied/hassled other pupils at school by hitting, kicking, or shoving them?

Victimization scale

This school year how often have you been . . .

1. Bullied/hassled by other students at school?

2. Bullied/hassled by other students at school by teasing them?

3. Bullied/hassled by other pupils at school by isolating them or shutting them out from others?

4. Bullied/hassled by other pupils at school by hitting, kicking, or shoving them?

The original instrument was independently translated from English into Italian by two authors of the present article. A third author combined these two translations into a single version. This single version was then back-translated independently by two PhD students (fluent in English as well as in Italian). The back-translated versions were almost identical to the original instrument (probably also due to the simple content of the items). The scales were pretested with a convenience sample of school students (n = 23) who had no apparent difficulty in understanding the meaning of the items.

Data Analysis

Data analyses were carried out in four steps using Mplus 7 (Muthén & Muthén, 1998-2012). Given that our data were obtained by cluster sampling and had a hierarchical structure (i.e., students are nested within classes), we used the “Type = complex” analytical approach and the Robust Maximum Likelihood (MLR) estimator. This approach, specially developed for analyzing complex survey data, makes it possible to compute standard errors and a chi-square test of model fit while taking into consideration the nonindependence of observations due to cluster sampling (Asparouhov, 2005; Asparouhov & Muthén, 2005). The very small amount of missing data (ranging from 0.6 to 0.9) was handled using the Full Information Maximum Likelihood method as implemented in Mplus (Muthén & Muthén, 1998-2012).

The sample was randomly divided into two subsamples: Sample 1 consisting of 13,451 students, and Sample 2 consisting of 13,219 students (Table 2). In a preliminary phase, an Exploratory Factor Analysis (EFA) was conducted on Sample 1 using Geomin oblique rotation (Yates, 1987). This rotation was chosen because it has been proven to work very well for simple and moderately complicated loading matrix structures (Asparouhov & Muthén, 2009). The decision on the number of factors to be retained was based on the criteria proposed by Preacher, Zhang, Kim, and Mels (2013), which suggest opting for a solution that best balances the desirable characteristics of parsimony and fit to observed data, and on the results of the parallel analysis (Horn, 1965) with the 95th percentile criterion (Glorfeld, 1995).

Table 2.

Characteristics of the Full Sample, Subsample 1, and Subsample 2.

Characteristic	Full sample (N = 26,670)	Subsample 1 (n = 13,451)	Subsample 2 (n = 13,219)
Average age (SD)	10.36 (.56)	10.39 (.53)	10.32 (.52)
Percentage of first-generation immigrants	3.3	3.5	2.9
Percentage of second-generation immigrants	6.5	6.8	5.4

In the second step, a confirmatory factor analysis (CFA) was carried out on Sample 2 and the model fit was assessed using MLR chi-square test statistic and multiple fit indices (comparative fit index [CFI], root mean square error of approximation [RMSEA], standardized root mean square residual [SRMR]) referring to common guidelines for an acceptable model fit (Hu & Bentler, 1999; Schreiber, Nora, Stage, Barlow, & King, 2006). In line with the theoretical expectations (Fandrem et al., 2009; Idsøe, Dyregrov, & Idsoe, 2012), we tested a model consisting of two correlated factors (Bullying and Victimization), in which each item had a nonzero loading on the factor that it was designed to measure and a zero loading on the other factor and error terms associated with each item were uncorrelated, with the exception of the two items about physical bullying (i.e., bullying others physically and being bullied physically). On the basis of previous studies which showed reciprocal effects between items specific to a physical domain (Idsøe et al., 2012; Marsh et al., 2011), the residuals of the two items about physical bullying were covaried.

In the third phase, the measurement invariance of the scales across immigrant background (native, first-generation immigrant, second-generation immigrant) was examined at the configural, metric, and scalar levels, by means of a hierarchical series of multigroup CFAs on the whole sample, imposing increasingly restrictive equality constraints on the model’s parameters (van de Schoot, Lugtig, & Hox, 2012). In each step of the analysis, the fit of the nested models was compared using two tests (Cheung & Rensvold, 2002): The corrected chi-square difference test and the change in CFI values (ΔCFI ≤ .01). Finally, the latent mean differences across immigrant status were estimated for each of the two scales.

Results

Table 3 shows bivariate correlations between the items of the Bullying and the Victimization scales together with the means and the standard deviations. The percentage frequency distribution within each item of the scales for all students, natives, first-generation, and second-generation immigrants are shown in Table 4. The mean scores and standard deviations of these groups of students on the Bullying and Victimization scales are summarized in Table 5.

Table 3.

Bivariate Correlations, Means, and Standard Deviations of the Items of the Bullying and Victimization Scales.

	Item 1	Item 2	Item 3	Item 4	Item 5	Item 6	Item 7	Item 8
Item 1	—
Item 2	.509	—
Item 3	.326	.326	—
Item 4	.317	.406	.273	—
Item 5	.236	.217	.145	.189	—
Item 6	.224	.306	.151	.239	.647	—
Item 7	.120	.164	.154	.159	.462	.470	—
Item 8	.198	.257	.180	.398	.340	.396	.314	—
M	1.57	1.39	1.46	1.23	1.95	1.69	1.59	1.24
SD	.615	.592	.624	.528	.817	.793	.755	.550

Note. All the correlations are statistically significant with p < .001.

Table 4.

Percentage Frequency Distributions Within Each Item of the Questionnaire for All Students, Natives, First-Generation, and Second-Generation Immigrants.

Response category	Item 1	Item 2	Item 3	Item 4	Item 5	Item 6	Item 7	Item 8
All students
Never	47.7	65.8	59.0	81.4	28.2	46.7	53.5	80.8
Now and then	48.9	31.1	37.3	15.8	56.2	42.5	38.1	15.9
Weekly	1.7	1.8	2.1	1.5	7.9	5.9	4.5	1.8
Daily	1.7	1.3	1.6	1.2	7.7	5.0	4.0	1.4
Natives
Never	47.9	66.2	58.8	82.6	28.8	47.6	54.5	81.8
Now and then	49.0	30.9	37.8	15.0	56.2	42.1	37.6	15.4
Weekly	1.6	1.7	1.9	1.4	7.7	5.7	4.2	1.7
Daily	1.5	1.1	1.5	1.0	7.3	4.6	3.7	1.2
First-generation immigrants
Never	48.0	66.5	65.8	76.4	25.6	42.9	50.5	77.6
Now and then	47.5	29.3	29.4	19.2	54.9	41.3	38.7	17.6
Weekly	2.3	3.0	3.2	1.9	8.7	7.6	5.7	2.3
Daily	2.2	1.2	1.7	2.4	10.7	8.1	5.0	2.5
Second-generation immigrants
Never	45.9	63.7	59.9	79.0	23.8	43.0	49.9	79.2
Now and then	50.7	34.1	36.6	18.3	60.5	46.4	41.5	17.8
Weekly	1.7	1.2	2.2	1.4	8.2	5.4	4.9	1.7
Daily	1.7	1.0	1.3	1.3	7.5	5.1	3.8	1.4

Table 5.

Means and Standard Deviations on the Bullying and Victimization Scales for All Students, Natives, First-Generation, and Second-Generation Immigrants (With Values of the Scales Ranging From 0 = Never to 3 = Daily).

	All students		Natives		First-generation immigrants		Second-generation immigrants
Scale	M	SD	M	SD	M	SD	M	SD
Bullying	0.41	0.42	0.40	0.41	0.42	0.47	0.42	0.41
Victimization	0.62	0.56	0.60	0.55	0.70	0.63	0.65	0.54

The results of the EFA showed that the solution with one factor had poor fit: χ²(20) = 5,309.713, p < .001; CFI = .678; RMSEA = .140 (90% confidence interval [CI] = [.137, .143]); SRMR = .097. The solution with two factors had a good fit to the observed data: χ²(13) = 1,598.367, p < .001; CFI = .949; RMSEA = .068 (90% CI = [.067, .075]); SRMR = .032. The factor loadings (Table 6) revealed a simple structure: All the items of the Bullying scale had high loadings on the first factor and very low loadings on the second factor, while the loadings of the items of the Victimization scale showed an inverse pattern. The correlation between the two factors was .405 (p < .05). The fit of the three-factor solution was χ²(7) = 242.837, p < .001; CFI = .986; RMSEA = .050 (90% CI = [.045, .056]); SRMR = .012. In this three-factor solution, the first two dimensions showed a pattern of factor loadings very similar to the two-factor solution (Table 6), while the third dimension had high loadings only on the two items about physical bullying (i.e., Item 4: bullying others physically, and Item 8: being bullied physically), which are known in the literature to show reciprocal effects (Idsøe et al., 2012; Marsh et al., 2011). These two items also had loadings higher than .30 on the first and on the second factor. Therefore, out of the three competing EFA solutions, the two-factor solution had the best balance between the two desirable characteristics of parsimony and good fit to the observed data (Preacher et al., 2013), offered a cleaner factor structure, and was consistent with the theoretical background of the scales. The results of the parallel analysis showed that the first three eigenvalues from the sample data were 3.064, 1.416, and 0.860. The 95th percentile eigenvalues derived from random data were 1.047, 1.032, and 1.020. Therefore, the parallel analysis corroborated the retention of two factors: Only the first two eigenvalues generated from the sample data were greater than those generated from parallel analysis. The two subscales had an acceptable internal consistency within Sample 1, with Cronbach’s alpha values of .70 for Bullying, and .76 for Victimization.

Table 6.

Exploratory Factor Analysis Results: Factor Loadings and Communalities.

	Two-factor solution			Three-factor solution
	Factor loadings			Factor loadings			Communalities
Items	Factor 1	Factor 2	Communalities	Factor 1	Factor 2	Factor 3	Communalities
Item 1	.654	−.010	0.423	.736	−.014	−.024	0.523
Item 2	.747	.002	0.559	.648	.016	.150	0.512
Item 3	.467	−.003	0.217	.426	−.010	.107	0.216
Item 4	.528	.066	0.312	.309	−.002	.596	0.553
Item 5	−.011	.787	0.612	.003	.826	−.113	0.640
Item 6	.054	.800	0.678	.019	.802	−.002	0.657
Item 7	−.010	.590	0.344	−.079	.598	.027	0.333
Item 8	.242	.372	0.270	−.002	.362	.441	0.421

Note. Factor loadings >.30 are in boldface.

The results of the CFA are presented in Figure 1. Except for the chi-square test (probably affected by the large size of the sample used in the present study) all the fit indices indicated a good fit of the model with the empirical data (Hu & Bentler, 1999; Schreiber et al., 2006): χ²(18) = 989.790, p < .001; CFI = 969; RMSEA = .045 (90% CI = [.043, .047]); SRMR = .031. All the factor loadings were statistically significant (p < .001). As expected, Bullying and Victimization were significantly correlated with each other (.461; p < .001). The two subscales showed acceptable internal consistency within Sample 2, with Cronbach’s alpha values of .70 for Bullying, and .76 for Victimization. In addition, the reliability of the scales was estimated on 30,000 bootstrap samples by means of coefficient omega (Dunn, Baguley, & Brunsden, 2014; Kelley & Pornprasertmanit, 2016). The results of this analysis confirmed the internal consistency of the scales, with a categorical omega value of .70 (95% CI = [.69, .71]) for Bullying and of .78 (95% CI = [.77, .80]) for Victimization within Sample 2, and categorical omega value of .71 (95% CI = [.69, .72]) for Bullying and of .77 (95% CI = [.76, .79]) for Victimization within Sample 1.

Figure 1.

Confirmatory factor analysis results.

In the multigroup CFAs, the comparison of the configural invariance model with the model with all the factor loadings constrained to be equal cross groups (metric invariance model) showed that the difference in the CFI between the models was smaller than the cut-off criterion (ΔCFI = .001), but the chi-square difference test was statistically significant (Δχ² = 34.253, p < .01). Given that this test could be not reliable with a large sample size and there was no substantial difference in CFI, we concluded that the hypothesis of metric invariance of the scales across immigrant background could be retained. The results of the comparison of the metric invariance model with the model in which all the item intercepts were constrained to be equal across groups (scalar invariance model) showed the full scalar invariance of the scales. In fact, the change in the CFI value was smaller than the cut-off criterion (ΔCFI = .004) and provided empirical support for the scalar invariance hypothesis, although the chi-square difference test was statistically significant (Δχ² = 73.558, p < .01).

Finally, Table 7 shows the results of the analysis of latent mean differences across immigrant status. Both first-generation and second-generation immigrant students proved to have higher levels of victimization than native students (the effect size in terms of Cohen’s d was, respectively, .21 and .10). Victimization proved to be lower in second-generation students than in first-generation students (Cohen’s d = .11). There were no statistically significant differences between the three groups on bullying.

Table 7.

Results of the Latent Factor Mean Differences Tests.

Latent factor	Difference tests
Latent factor	Natives vs. first-generation immigrants^a	Natives vs. second-generation immigrants^a	First-generation vs. second-generation immigrants^b
Bullying	.018	.021	.002
Victimization	.123**	.058**	−.065*

The natives are the reference group for the comparison (the latent mean for this group is fixed to be zero).

The first-generation immigrants are the reference group for the comparison (the latent mean for this group is fixed to be zero).

p < .05. **p < .01.

Discussion

In the present study, we examined the factor structure of a widespread self-report measure of bullying and victimization as well as the measurement invariance of this measure across groups of students with different migration backgrounds. On the basis of a nationally representative sample of Italian fifth-grade students, the rates of prevalence of bullying and victimization were then estimated for first-generation immigrants, second-generation immigrants and native students.

The results of the psychometric analysis showed that the factor structure of the instrument in Italy was much as expected on the basis of previous studies conducted in other cultural contexts (Fandrem et al., 2009; Idsøe et al., 2012; Tiliouine, 2014), revealing two related but separate factors. The reliability of the scales in terms of their internal consistency was acceptable (.70 for Bullying and .76 for Victimization; Nunnally & Bernstein, 1994). The scales showed complete metric and scalar invariance across immigrant backgrounds, and this supported the comparison of latent means between the groups of students. The results indicated that the latent means for first-generation and second-generation immigrants in the victimization scale were lower than for natives, while there were no significant differences in the bullying scale. This suggests that in Italy immigrant pupils are, in general, at a higher risk of being victimized by their peers. These findings are consistent with results obtained by Graham and Juvonen (2002) and by Strohmeier et al. (2011) in other cultural contexts. Our results also showed that, although second-generation students have a higher probability of being bullied than native students, this risk is significantly lower than in the case of first-generation immigrant students. This finding seems to support the hypothesis that differences between natives and immigrants may become attenuated over time as one generation is followed by the next (Barban & White, 2011; Neidert & Farley, 1985). The effect sizes of the latent mean differences on victimization were between .2 and .1, which means they are in line with previous studies in the European context that revealed differences between natives and immigrant students (Strohmeier et al., 2011). These findings indicated that the magnitude of the statistically significant differences as regards the probability of being victimized by peers between native students, first-generation, and second-generation immigrants was small but meaningful and nontrivial (Cohen, 1988).

The results of the present study have implications for dealing with bullying in multicultural contexts. We showed that immigrant students are more frequently the victims of bullying, and other studies provided evidence that these pupils appear to be more socially isolated in the classroom (Alivernini & Manganelli, 2016), as well as more oriented toward competition at school than their peers (Alivernini et al., 2018). This suggests that interventions at school should aim to create networks of social support for immigrant children. An interesting possibility for intervention is that of peer mentoring, which has proved to be quite effective in increasing a sense of connection with school and cohesion between students (Vreeman & Carroll, 2007). As Newman, Holden, and Delville (2005) have pointed out, this type of mentoring provides social support at the same time as being an effective way to model nonviolent behaviors. An important question for future research is whether interventions should consider issues related to students’ ethnicity. Unfortunately, information about different ethnic groups was not available in our database and, while most immigrant students in Italian primary schools are Caucasian and from Europe, there are still various cultural differences that would be worth studying. In addition, future research will be necessary to generalize the results of the present study to other grades, as it was based on a representative sample of fifth-grade students.

In conclusion, in spite of the limitations we have mentioned, we believe that the present study has contributed to the literature in several ways. We have expanded the repertory of instruments available for the child population by adapting two short scales validated for measuring bullying and victimization, thereby making it possible to conduct international comparisons with Italy, a country of particular interest for the study of these behaviors in multicultural schools. We have also addressed the problem of the measurement invariance of self-report measures of bullying and victimization across various immigration backgrounds. Many self-report measures feature questions which are the same, or very similar to, those included in the tested instrument (Ramelow, Currie, & Felder-Puig, 2015), and we have provided some initial evidence to support the claim that these items have much the same meaning across groups of students with different countries of origin. Finally, on the basis of a representative sample of primary school students and taking measurement errors into consideration, we have provided prevalence estimates among immigrant and native primary school students of bullying and victimization, two problems which seriously increase the risk of psychosocial dysfunction (Evans-Lacko et al., 2017; Hellfeldt, Gill, & Johansson, 2018; Mooij, 2012).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Alivernini

Manganelli

(2011). Is there a relationship between openness in classroom discussion and students’ knowledge in civic and citizenship education? Procedia–Social and Behavioral Sciences, 15, 3441-3445. doi:10.1016/j.sbspro.2011.04.315

Alivernini

Manganelli

(2016). The classmates social isolation questionnaire (CSIQ): An initial validation. European Journal of Developmental Psychology, 13, 264-274. doi:10.1080/17405629.2016.1152174

Alivernini

Manganelli

Lucidi

(2016). The last shall be the first: Competencies, equity and the power of resilience in the Italian school system. Learning and Individual Differences, 51, 19-28. doi:10.1016/j.lindif.2016.08.010

Alivernini

Manganelli

Lucidi

(2018). Personal and classroom achievement goals: Their structures and relationships. Journal of Psychoeducational Assessment, 36, 354-365.

Asparouhov

(2005). Sampling weights in latent variable modeling. Structural Equation Modeling, 12, 411-434.

Asparouhov

Muthén

(2005). Multivariate statistical modeling with survey data. Proceedings of the Federal Committee on Statistical Methodology (FCSM) Research Conference. Retrieved from http://www.fcsm.gov/events/papers05.html

Asparouhov

Muthén

(2009). Exploratory structural equation modeling. Structural Equation Modeling, 16, 397-438.

Baldry

A. C.

(1998). Bullying among Italian middle school students. School Psychology International, 19, 361-374.

Barban

White

M. J.

(2011). Immigrants’ children’s transition to secondary school in Italy. International Migration Review, 45, 702-726. doi:10.1111/j.1747-7379.2011.00863.x

10.

Berry

J. W.

(2001). A psychology of immigration. Journal of Social Issues, 57, 611-627.

11.

Berry

J. W.

Phinney

J. S.

Sam

D. L.

Vedder

(2006). Immigrant youth: Acculturation, identity, and adaptation. Applied Psychology: An International Review, 55, 303-332.

12.

Berry

J. W.

Poortinga

Y. H.

Segall

M. H.

Dasen

P. R.

(2002). Cross-cultural psychology: Research and applications (2nd ed.). Cambridge, UK: Cambridge University Press.

13.

Boulton

M. J.

(1995). Patterns of bully/victim problems in mixed race groups of children. Social Development, 4, 277-293.

14.

Cavicchiolo

Alivernini

Manganelli

(2016). Immigrants are like . . . the representation of immigrants in Italy: The metaphors used by students and their family backgrounds. Journal of Educational, Cultural and Psychological Studies, 13, 161-188.

15.

Cheung

G. W.

Rensvold

R. B.

(2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9, 233-255.

16.

Cohen

(1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

17.

Dunn

T. J.

Baguley

Brunsden

(2014). From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation. British Journal of Psychology, 105, 399-412. doi:10.1111/bjop.12046

18.

Eslea

Mukhtar

(2000). Bullying and racism among Asian schoolchildren in Britain. Educational Research, 42, 207-217.

19.

Evans-Lacko

Takizawa

Brimblecombe

King

Knapp

Maughan

Arseneault

(2017). Childhood bullying victimization is associated with use of mental health services over five decades: A longitudinal nationally representative cohort study. Psychological Medicine, 47, 127-135. doi:10.1017/S0033291716001719

20.

Fandrem

Ertesvåg

S. K.

Strohmeier

(2010). Bullying and affiliation: A study of peer groups in native and immigrant adolescents in Norway. European Journal of Developmental Psychology, 7, 401-418. doi:10.1080/17405620802335075

21.

Fandrem

Strohmeier

Roland

(2009). Bullying and victimization among native and immigrant adolescents in Norway: The role of proactive and reactive aggressiveness. The Journal of Early Adolescence, 29, 898-923. doi:10.1177/0272431609332935

22.

Foundation for Initiatives and Studies on Multi-Ethnicity. (2015). Alunni con cittadinanza non italiana –Tra difficoltà e successi (Rapporto nazionale a.s. 2013/2014) [Students without Italian citizenship–Difficulties and successes] (National Report 2013/2014, Quaderniismu No. 1). Retrieved from http://www.ismu.org/2015/02/alunni-con-cittadinanza-non-italiana-presentati-dati/

23.

Land

K. C.

Lamb

V. L.

(2012). Bullying victimization, socioeconomic status and behavioral characteristics of 12th graders in the United States, 1989 to 2009: Repetitive trends and persistent risk differentials. Child Indicators Research, 6, 1-21. doi:10.1007/s12187-012-9152-8

24.

Genta

M. L.

Menesini

Fonzi

Costabile

Smith

P. K.

(1996). Bullies and victims in schools in central and southern Italy. European Journal of Psychology of Education, 11, 97-110. doi:10.1007/BF03172938

25.

Glorfeld

L. W.

(1995). An improvement on Horn’s parallel analysis methodology for selecting the correct number of factors to retain. Educational and Psychological Measurement, 55, 377-393. doi:10.1177/0013164495055003002

26.

Graham

Juvonen

(2002). Ethnicity, peer harassment, and adjustment in middle school: An exploratory study. The Journal of Early Adolescence, 22, 173-199. doi:10.1177/0272431602022002003

27.

Hellfeldt

Gill

P. E.

Johansson

(2018). Longitudinal analysis of links between bullying victimization and psychosomatic maladjustment in Swedish schoolchildren. Journal of School Violence, 17, 86-98.

28.

Horn

J. L.

(1965). A rationale and test for the number of factors in factor analysis. Psychometrika, 30, 179-185.

29.

Bentler

P. M.

(1999). Cut-off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

30.

Idsøe

Dyregrov

Idsoe

E. C.

(2012). Bullying and PTSD Symptoms. Journal of Abnormal Child Psychology, 40, 901-911. doi:10.1007/s10802-012-9620-0

31.

Juvonen

Graham

(2014). Bullying in schools: The power of bullies and the plight of victims. Annual Review of Psychology, 65, 159-185. doi:10.1146/annurev-psych-010213-115030

32.

Kelley

Pornprasertmanit

(2016). Confidence intervals for population reliability coefficients: Evaluation of methods, recommendations, and software for composite measures. Psychological Methods, 21, 69-92. doi:10.1037/a0040086

33.

Marsh

H. W.

Nagengast

Morin

A. J. S.

Parada

R. H.

Craven

R. G.

Hamilton

L. R.

(2011). Construct validity of the multidimensional structure of bullying and victimization: An application of exploratory structural equation modeling. Journal of Educational Psychology, 103, 701-732. doi:10.1037/a0024122

34.

McKenney

K. S.

Pepler

Craig

Connolly

(2006). Peer victimization and psychosocial adjustment: The experiences of Canadian immigrant youth. Electronic Journal of Research in Educational Psychology, 9, 230-264.

35.

Midthassel

Bru

Idsoe

(2008). Is the sustainability of reduction in bullying related to follow-up procedures? Educational Psychology, 28, 83-95.

36.

Milfont

T. L.

Fischer

(2015). Testing measurement invariance across groups: Applications in cross-cultural research. International Journal of Psychological Research, 3, 111-130.

37.

Mooij

(2012). A Mokken scale to assess secondary pupils’ experience of violence in terms of severity. Journal of Psychoeducational Assessment, 30, 496-508. doi:10.1177/0734282912439387

38.

Moran

Smith

P. K.

Thompson

Whitney

(1993). Ethnic differences in experiences of bullying: Asian and White children. British Journal of Educational Psychology, 63, 431-440.

39.

Muthén

L. K.

Muthén

B. O.

(1998-2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

40.

National Institute for the Evaluation of the Education System. (2011). La terza indagine IEA sull’Educazione Civica e alla Cittadinanza. Rapporto Nazionale. [The third IEA survey on Civic and Citizenship Education. National Report]. Author. Retrieved from http://www.invalsi.it/download/rapporti/iccs2009/Rapporto_ICCS_2009.pdf

41.

National Institute for the Evaluation of the Education System. (2014). Rilevazioni Nazionali sugli Apprendimenti 2013-2014 [National Evalutation of Learning]: Author. Retrieved from http://www.invalsi.it/areaprove/rapporti/Rapporto_Rilevazioni_Nazionali_2014.pdf

42.

Neidert

Farley

(1985). Assimilation in the United States: An analysis of ethnic and generation differences in status and achievement. American Sociological Review, 50, 840-850.

43.

Newman

M. L.

Holden

G. W.

Delville

(2005). Isolation and the stress of being bullied. Journal of Adolescence, 28, 343-357. doi:10.1016/j.adolescence.2004.08.002

44.

Nunnally

J. C.

Bernstein

I. H.

(1994). Psychometric theory. New York, NY: McGraw-Hill.

45.

Olweus

(1994). Bullying at school: Basic facts and effects of a school based intervention program. Journal of Child Psychology and Psychiatry, 35, 1171-1190. doi:10.1111/j.1469-7610.1994.tb01229.x

46.

Organisation for Economic Co-Operation and Development. (2014). PISA 2012 technical report. Programme for International Student Assessment, Organisation for Economic Co-Operation and Development Publishing. Retrieved from http://www.oecd.org/pisa/pisaproducts/PISA-2012-technical-report-final.pdf

47.

Preacher

K. J.

Zhang

Kim

Mels

(2013). Choosing the optimal number of factors in exploratory factor analysis: A model selection perspective. Multivariate Behavioral Research, 48, 28-56.

48.

Ramelow

Currie

Felder-Puig

(2015). The assessment of school climate: Review and appraisal of published student-report measures. Journal of Psychoeducational Assessment, 33, 731-743. doi:10.1177/0734282915584852

49.

Roland

Idsøe

(2001). Aggression and bullying. Aggressive Behavior, 27, 446-462. doi:10.1002/ab.1029

50.

Schreiber

J. B.

Nora

Stage

F. K.

Barlow

E. A.

King

(2006). Reporting structural equation modeling and confirmatory factor analysis results: A review. Journal of Educational Research, 99, 323-337.

51.

Strohmeier

Kärnä

Salmivalli

(2011). Intrapersonal and interpersonal risk factors for peer victimization in immigrant youth in Finland. Developmental Psychology, 47, 248-258. doi:10.1037/a0020785

52.

Strohmeier

Spiel

(2003). Immigrant children in Austria: Aggressive behavior and friendship patterns in multicultural school classes. Journal of Applied School Psychology, 19, 99-116.

53.

Tiliouine

(2014). School bullying victimisation and subjective well-being in Algeria. Child Indicators Research, 8, 133-150. doi:10.1007/s12187-014-9286-y

54.

van de Schoot

Lugtig

Hox

(2012). A checklist for testing measurement invariance. European Journal of Developmental Psychology, 9, 486-492.

55.

Verkuyten

Thijs

(2002). Racist victimization among children in The Netherlands: The effect of ethnic group and school. Ethnic and Racial Studies, 25, 310-331.

56.

Verkuyten

Thijs

(2006). Ethnic discrimination and global self-worth in early adolescents: The mediating role of ethnic self-esteem. International Journal of Behavioral Development, 30, 107-116.

57.

Vieno

Lenzi

Gini

Pozzoli

Cavallo

Santinello

(2015). Time trends in bullying behavior in Italy. Journal of School Health, 85, 441-445. doi:10.1111/josh.12269

58.

Vreeman

R. C.

Carroll

A. E.

(2007). A systematic review of school-based interventions to prevent bullying. Archives of Pediatrics & Adolescent Medicine Journal, 161, 78-88. doi:10.1001/archpedi.161.1.78

59.

Woods

Wolke

(2004). Direct and relational bullying among primary school children and academic achievement. Journal of School Psychology, 42, 135-155. doi:10.1016/j.jsp.2003.12.002

60.

Yates

(1987). Multivariate exploratory data analysis: A perspective on exploratory factor analysis. Albany: State University of New York Press.