Family Acculturation in Host and Immigrant Couples: Dyadic Research in an Italian Context

The purpose of this research is to study acculturation strategies and attitudes in central and peripheral domains of host and immigrant couples in an Italian context. The participants were 60 dyads (30 host couples and 30 immigrant couples) who completed a questionnaire based on the Relative Acculturation Extended Model (RAEM). Based on the analysis, we found that the general acculturation attitude preferred by immigrant couples is integration, and Italian couples prefer that immigrants adopt it. Furthermore, Italian partners show moderate internal agreement, whereas immigrant couples show a high degree of agreement. In both groups, the level of agreement between dyadic members is only partially determined by their membership within a social group. The socio-cultural context has a significant role in the internal similarity of Italian couples. In contrast, there is dyadic agreement within immigrant couples.

which the present study intends to investigate. Therefore, in this article, we build on previous research in this area in four significant ways. First, we study the acculturation process at the family level, by focusing on couple dyads. Second, we adhere to the current views of the two-dimensional acculturation model, and then we apply this perspective on a dyadic level by considering the parental couple. Third, we use the dyadic level analysis of a parenting couple in the migration process and compare natives and immigrants. Finally, we emphasize that there are different ways of acculturation with respect to central life domains and peripheral life domains and therefore there could be different agreements within couples and consequently at the family level.

Aims
The aims of the present research are to study the acculturation process of Italian and immigrant families. Specifically, this study analyzes general acculturation attitudes and strategies and attitudes in life domains using a family approach in host and immigrant couple dyads (married or cohabiting) with children in early childhood.
The aims of the present study are as follows: 1. to describe the general acculturation attitudes and life domain strategies and attitudes adopted by immigrant dyads and the perceptions of host dyads regarding the strategies and attitudes adopted by immigrant dyads; 2. to analyse the internal dyadic accord in Italian and immigrant couples with respect to acculturation strategies and attitudes; 3. to verify if the dyadic agreement between the two members of a couple is effective and not determined by membership of the social groups of women and men; 4. to study if the dyadic agreement diminishes once the stereotypical effect has been controlled for, because the socio-cultural context has a significant role in the internal similarity of the dyads.

The Italian Context
Based on the migration flows towards Italy in recent years, the number of immigrants has greatly increased.
The regular resident population in Italy in 2014, according to ISTAT, included almost 5 million foreigners. Over the past 10 years according to the results of the 15th general census of 2011 (ISTAT, 2013), the Italian population increased by 2.7 million people. This phenomenon appears to be attributable to the increase in foreign nationals to 5,029,000 as of January 1, 2017, equal to 9.05% of the total Italian population, which is estimated to be 60,579,000. The ISTAT (2017) report also shows that a change occurred in the last twenty years in migration flows in terms of the heterogeneity of backgrounds and the stabilization of immigrant presence in the territory.
Indicators of this phenomenon are the increase in indefinite resident permits and acquisition of citizenship by naturalization, specifically, transmitted from parents to children and reaching adult age for those born in Italy.
The statistics also report the strong tendency of immigrants to recompose their family of origin in Italy since they often have children who were previously born in the country of origin and decide to extend the family by having children in Italy. Furthermore, recent statistics on Italian family reunions state that most of the migratory movements are tied to reasons concerning the family.

Measures
Among the different acculturation models present in the literature, it was decided to use the Relative Acculturation Extended Model (RAEM; Navas Luque, Garcia, & Rojas, 2006;Navas Luque, Rojas, Garcia, & Pumares, 2007;, which addresses both acculturation strategies and acculturation attitudes.
The acculturation strategies consider the real plan/situation for each domain, specifically the acculturation options that immigrants say they put into practice within the new society versus those that the natives perceive as being put into practice by immigrants. The acculturation attitudes refer to the ideal plan, that is, the acculturation options that immigrants would choose to put into practice if they could and that natives would prefer immi- The combination of the questions presented in

Procedure
The participants were contacted through their children's teachers, and a letter described the aims of the study.
Then, they met the researchers and completed the self-report questionnaire. An assistant researcher was present while the questionnaire was being completed to supervise data collection. Instructions were given at the start of the study informing people that participation was voluntary, and that responses were confidential and anonymous in compliance with the Italian law on privacy (No.196/2003). The procedures fully complied with the research ethical code of the Italian Association of Psychology, and the informed consent protocol was provided to participants.

Data Analysis
Data were analysed using SPSS software for statistical analysis, version 18. To describe the types of general acculturation attitudes, mean scores and standard deviations were calculated. To integrate individuals into one of four general strategies, in line with the work of Mancini and Bottura (2014) and Navas Luque et al. (2010), single sample t-tests were run for each answer using the average value of 3. Scores below 3 for both responses indicated the adoption of a strategy/attitude of marginalization; if the score for the first question was higher than 3 and the score for the second question was lower than 3, the strategy/attitude was separation; if the score for the first question was less than 3 and the score for the second question was higher than 3, the strategy/attitude was assimilation; if the score for both questions was higher than 3, a strategy of integration was Family Acculturation in Host and Immigrant Couples 918 indicated; if the score for one or both questions was equal to 3, the responses were categorized as indeterminate, as proposed by Navas Luque et al. (2010). To describe acculturation strategies and attitudes for domains in Italian and immigrant dyads, average scores and standard deviations were calculated. The data analysis of dyads includes several techniques that enable considering the complexity of family relationships and their interdependence. Lanz and Rosnati (2002) argue that in the literature, there is no score that represents the complexity of family relationships. For this reason, researchers usually use multiple dyadic indices to compare them (Fisher, Kokes, Ransom, Philips, & Rudd, 1985). The differences between host and immigrant dyads were calculated by independent sample t-tests. Furthermore, paired samples t-tests to analyse the internal differences of the members of the dyads of both groups were performed. The effect size was described with Cohen's d coefficient.
The dyadic correlations (r dyadic ) and the index discrepancy (d) were also used. Dyadic correlations were used to detect the degree of similarity within the dyad (Lanz & Rosnati, 2002), considering the preservation of the culture of origin and the adoption or maintenance of the host culture in peripheral and central domains. The discrepancy index was calculated to detect the level of disagreement between the two members of the dyad on the same constructs. For dyadic correlations, a correlation index was applied, according to Alfieri and Marta (2012). The correlation index used in the present work was Pearson's, which enables verifying if the constructs are specific to the social groups to which they belong (women and men) or to dyads. In fact, a significant correlation indicates that a particular construct is specific to the social groups and not to the dyad (women and men of the same pair); conversely, a low correlation indicates the specificity of that size to the dyad.
To measure the stereotypical effect, the following procedure proposed by Kenny and Acitelli (1994) was applied: the average of the responses of all husbands (MH1) on the same item or that of wives (MW1) on the same item was subtracted from the husband's response (H1) and the wife's response (W1) to each individual item of a scale, respectively. Therefore, MH1 and MW1 represented the stereotypes of the response of the husbands and the wives, respectively. Before calculating the dyadic correlation, we subtracted MH1 or MW1 from the item. The differences between the scores of the Italian and immigrant dyads on the scale were calculated by independent samples t-test. The effect size was described with Cohen's d coefficient. Comparisons between the average dyadic scores (r dyadic ) and dyadic cleaned scores (r "pure" ) were calculated using paired samples ttests. Figure 1 shows the scores on general acculturation attitudes that, according to the Italian dyads, are adopted by immigrants and those that immigrant dyads claim to adopt. The figure was generated by combining the scores on the two main questions described above in the instrument section according to Berry's model (1980).

Results
Specifically, it is a two-dimensional model that describes the four acculturation outcomes: assimilation, integration, marginalization and separation.
Rania, Migliorini, & Rebora 919 Family Acculturation in Host and Immigrant Couples 920 To verify the level of agreement between the partners of the dyads in relation to the investigated variables, we calculated dyadic correlations (Lanz & Rosnati, 2002). Table 3 shows the scores of internal agreement with the dyads (r dyadic ), the discrepancy index (d) and the Pearson correlation.
Regarding the real plan for the Italian dyads, the level of agreement between the members was moderate, with the exception of the perception of maintenance of culture of origin by immigrants in peripheral domains, which had a high amount of dyadic agreement, and with the exception of the perception of maintaining the culture of origin by immigrants in central domains, for which the level of dyadic agreement was very low. Therefore, the discrepancy index was reported to be moderate although it trended towards 1, which indicated a lack of internal agreement in dyads.
The difference between the dyadic relationship and the index of discrepancy scores was attributable to differences between the mean scores and the standard deviations between the members of dyads. The Pearson correlations of the perception of Italians maintenance of the culture of origin by immigrants were moderate and sig-  Regarding the ideal plan, the dyadic correlation showed that there was a low level of agreement between partners in relation to the perception that immigrants desire to retain their culture of origin and to adopt the culture of the dominant society in the central domain, whereas there was good agreement regarding the perception that immigrants desire to maintain their culture and to adopt Italian culture in the peripheral domains. The values of the discrepancy index were high. In sum, the members of Italian dyads showed a level of agreement that ranged from moderate to low. Regarding the ideal plan of the desire of immigrants to maintain their culture in peripheral domains of life, the correlation between Italian women and men was strong and significant, indicating the specificity of the social groups (women and men) and not the dyads; however, the desire of immigrants to adopt Italian culture in the central domains of life and adopt Italian culture in the peripheral domains had a moderate and not significant relation. A weak and not significant relationship emerged concerning the desire to maintain the culture of origin in the central domains, which indicated a dyadic specificity with respect to this dimension. In general, the agreement between members was greater and more specific to dyads in the central domains of life than in the peripheral domains.
Immigrant dyads showed a good degree of agreement in the real and ideal plans; in fact, the correlations of the values of the dyads were high on all dimensions. Moreover, to confirm this, immigrant dyads had a discrepancy index at a medium level for the real plan, whereas for the ideal plan, the discrepancy index was closer to 1, indicating a trend towards agreement within dyads. Therefore, in general, partners within immigrant dyads seem to have a high level of agreement. As for the Pearson correlations of immigrant dyads, the real plan dimensions were all characterized by strong and significant relations, and with respect to the adoption of Italian culture in the peripheral domains of life, a moderate and significant relationship was found. Regarding the ideal plan, the trend of the correlation was the same; in fact, all dimensions had strong and significant correlations, with the exception of the preference to maintain Italian culture in the central domains of life, for which the relationship was moderate and significant. Consequently, from these results it seems that agreement between im-Family Acculturation in Host and Immigrant Couples 922 migrant women and men was primarily related to a social factor (belonging to groups of women and men). Only the maintenance of Italian culture in the central domains in the real plan was specific to the dyad. For the real plan, internal agreement between the members was high, and it was in the moderate range only in the case of the adoption of Italian culture in the central domains of life. However, for the ideal plan, the agreement was high in all dimensions, showing a high level of similarity between immigrant partners. To confirm that the level of agreement among members of a dyad was not largely due to the cultural context in which they live, we performed a comparison between the dyadic score and the dyadic score r "pure" , the score from which the stereotypical effect was subtracted (Table 4). The levels of internal similarity in the host dyads for the real plan were high compared to the perception of immigrants' maintenance of their culture in peripheral domains. With regard to the adoption of Italian culture by immigrants both in the peripheral and central domains, the levels of internal similarity were moderate; the relationship decreased considerably with respect to the perception of immigrants' maintenance of their culture in the central domains of life. For the ideal plan, the level of internal agreement of Italian dyads was high concerning the peripheral domains of life and whether they preferred immigrants to retain their culture or adopt the Italian culture; however, there was a moderate level of internal agreement regarding to the central domains of life both in the preference that immigrants adopt Italian culture and in the preference to maintain their own culture.
After calculating the stereotypical effect, the scores decreased for the real plan relative to the maintenance of their culture in peripheral domains of life, and in the adoption of Italian culture both in the peripheral and central domains. These results are characteristic of the social groups (host and immigrants) to which the dyads belong.
On the contrary, the score for maintaining the culture of origin in the central domains of life increased and was specific to the agreement between dyad partners; the scores decreased for the ideal plan with regard to the desire that immigrants adopt Italian culture and maintain their culture of origin in peripheral domains of life, thus being specific to the social group to which they belonged (host and immigrants). Conversely, the scores increased when compared to the desire that immigrants maintain their culture or adopt the Italian culture in the central domains. These results would be specific to the level of agreement between dyad partners. Rania,Migliorini,& Rebora 923 By carrying out the paired samples t-test between the dyadic correlations and the correlations and correcting for the stereotypical effects, significant differences emerged for the real plan regarding the dimension of the perception of immigrants' maintenance of their own culture in peripheral domains of life (t(25) = 3.47, p < .01).
Shifting the focus to the immigrant dyads, in regards to the real plan, the levels of similarity between the members of a dyad were high, except for the adoption of Italian culture in the central domains of life which was moderate. Regarding the ideal plan, agreement was high in all dimensions. After calculating the stereotypical effect for the real plan, the scores related to the maintenance of their culture and the adoption of Italian culture in the peripheral domains were lower. On the contrary, the scores increased relative to the maintenance of their culture and the adoption of Italian culture in the central domains, highlighting the specificity of these dimensions to the level of agreement between dyad partners. Instead, regarding the ideal plan, the scores were low for all dimensions. Although the scores for immigrant dyads decreased, they were still higher than .50 after the removal of the stereotypical effect, indicating high internal agreement for dyads for all the RAEM's dimensions.
Since there was significant agreement between the members, there were no significant differences when comparing the averages for paired samples between the dyadic correlation and the correlation corrected for the stereotypical effect.

Discussion and Conclusions
This study aimed to describe the general acculturation attitudes and acculturation strategies and attitudes in the peripheral and central domains of life from a family perspective in host and immigrant dyads. In line with previous studies conducted at the individual level (Berry, 1997;Navas Luque et al., 2006;Van Oudenhoven et al., 1998;Zagefka & Brown, 2002), the general acculturation attitude preferred by immigrant dyads was integration.
The Italian dyads, in line with the immigrant dyads, perceived that immigrants preferred the option of integration, although there was an evident tendency towards assimilation. Regarding the perceptions of the host culture, Zagefka, Tip, Gonzales, Brown, and Cinnirella (2012) explain how it is possible that local residents, perceiving the intention of immigrants to adopt their culture, are more willing to open up to them.
The present research offers a relevant contribution to the acculturation literature by exploring the acculturation practices of families in an intercultural context. The analysis of the dyadic dimension represents an important novelty in the studies on acculturation considering the close relationship between acculturation and family relationships, as the family influences the effects of migration on future generations. Namely, within the acculturation process, it is necessary to consider the contact within different family models that can influence one another in some dimensions of family life .
The present study also aimed to verify the existence of a dyadic agreement between the members of host couples and those of immigrant couples with respect to acculturation strategies and attitudes. Generally, the Italian dyads showed moderate internal agreement, whereas for immigrant dyads, a high degree of agreement between members was found. In fact, dyadic agreement could be considered support for the negotiation of acculturation strategies not only at the couple level but also at the family level (Kisselev, Brown, & Brown, 2010). The high degree of agreement in immigrant dyads could also be considered in line with the idea that the recognition and protection of family ties represent a focal point for immigrants (Regalia & Giuliani, 2014). As suggested by McGoldrick and Ashton (2012), migration is a factor that influences how family members relate to their cultural heritage, to others of their cultural group, and to the preservation of cultural traditions. However, a recent study Family Acculturation in Host and Immigrant Couples 924 by Migliorini, Rania, Tassara, and Cardinali (2016) points out that migrant families obtain lower scores than host dyads regarding ritual symbolic significance; this finding indicates that further studies are needed on this topic.
Another aim was to investigate whether the degree of dyadic agreement between the members of a couple was effective and not determined by membership of the social groups of men and women with respect to acculturation domains. In the present study, the Italian dyads showed a dyadic specificity regarding the central domains in life based on the real plan, and specificity for groups (women and men) regarding the peripheral domains. In relation to the ideal plan, dyadic specificity applied to the preference for adopting Italian culture both in the peripheral and central domains, but, for the central domains, dyadic specificity applied only to the preference that immigrants maintain their culture of origin. On the contrary, agreement regarding the preference of maintenance by immigrants of their culture of origin in the peripheral domains was specific to the groups.
Moreover, the immigrant dyads showed specific agreement between dyadic members only in the adoption of Italian culture in the peripheral domains of life, and agreement on all remaining dimensions seemed to be specific to the gender groups to which they belonged (men and women). According to Regalia and Giuliani (2014) these gendered differences are not uncommon and could be related to on various areas of life (values, practices, behaviours).
The final aim was to investigate whether the dyadic agreement would be decreased by controlling the stereotypical effect, specifically, whether the socio-cultural context played a significant role in the internal similarity of dyads, according to the methodology of dyadic analyses followed. Regarding the Italian dyads, after controlling for stereotypical effects on the dyadic agreement, what emerged was that the environment prevailed in defining the internal similarity between members. Immigrant dyads, however, did not seem to be influenced by the context, but the dyadic agreement was similar to the responses provided by dyad members. This finding could be explained from an intergroup perspective, underlining that coherence between individuals of a minority group would be essential to not lose their cultural specificity (Verkuyten & Martinovic, 2014).
An aspect that is not very thoroughly studied in the literature like the theme of family acculturation is a strong point of this study, it may prove to be a limitation because the lack of literature on the subject hinders an effective comparison with data from other similar research in terms of content and methodology. Furthermore, another possible limitation comes from the lack of consideration of the individual variables in determining host and immigrant acculturation preferences; these individual variables are presented in the literature, and include psychological issues, group and national identification (Mashuri, Burhan, & van Leeuwen, 2013;Nesdale & Mark, 2000;Verkuyten, & Martinovic, 2012) and, ethnic identity (Phinney, Horenczyk, Liebkind, & Vedder, 2001). Another limitation to this research is that, although the use of convenience sampling was adequate considering the explorative nature of this study (De Carlo & Robusto, 1996), it did not allow us to obtain a high degree of representativeness in the sample. For this reason, the results should be interpreted with caution as they cannot be generalized to the full range of host and migrant couples.
According to Kisselev, Brown, and Brown (2010), qualitative research is often culturally congruent and consequently could be an effectively used multimethod approach that allows us to understand and draw conclusions about immigrant families. Future research could more deeply investigate this issue by expanding the sample size and by considering other measures of individual, couple and family well being that could in some way affect the participants' dyadic agreement. Therefore, future studies should focus on the direct measurement of acculturation strategies and attitudes implemented and desired by the native population in the peripheral and Rania,Migliorini,& Rebora 925 central domains of life since in the present study, only their perception of acculturation processes affecting immigrants was assessed. Furthermore, researchers who are interested in the acculturation processes of parenting dyads should also consider sample those who have divorced or no longer share a partnership. These further investigations could lead to understanding if acculturation and couple agreement could be maintained when the relationship is broken. Moreover, it would be interesting for future research to consider immigrants who have lived in the host country different lengths of time, to understand the differences related to the length of residency. The immigrant participants in this study had lived in Italy for approximately ten years, which could be considered a long time during which they could have modified their level of dyadic acculturation agreement.
This research would enable adding another piece to the complex picture of Italian intercultural relationships.
With the increase in migratory flows and family reunification in various European countries, the scientific community should continue to study the processes of acculturation at the parental couple level to understand developments at the family level and understand how these dyadic agreements have repercussions at the generational level (Fine & Fincham, 2013). Until now, studies have focused on the dyadic level between parents and children. It is important to underline that successful integration could be considered a function of both acculturation strategies at the individual, dyadic and group levels in different domains, and the policies and practices of the receiving culture (Sabatier & Berry, 2008) in terms of multiculturalism and/or intergroup relationships.
Practical implications should take into account the different couples and family models that come into contact with the migration processes (Singh & Dutta, 2010). In fact, we believe our research has important strengths and interesting findings that could have practical implications for family-based interventions in community enrichment and school programs to support continuity in migrant family identity (Cigoli & Scabini, 2006) and to promote the well-being of both native and immigrant family members.

Funding
The authors have no funding to report.