Cohabitation among Second Generation Swedes of Turkish and Polish Origin

Eva Bernhardt, Stockholm University
Calvin Goldscheider, Brown University

This paper focuses on the transition to adulthood among second generation Swedes of Polish and Turkish origins, with a particular focus on cohabitation. Research on immigrants in various countries has documented the important role of family relationships as a resource for adaptation and as a source of family values. However, relationships between parents and their children and between union partners are challenged by immigration from societies with patriarchal families to societies with egalitarian family systems and social policies reinforcing gender equality, and the challenge is likely to be even greater in Sweden, with its high levels of cohabitation. Using data from national representative surveys in Sweden that include both attitudes toward cohabitation and actual partnership histories, we address the question of how much young adults of Turkish and Polish origins have become Swedish in their family relationships. We include, in addition to standard predictors of union formation, data collected on measures of ethnic intensity and attitudes toward inter-ethnic partnering. Sweden has experienced rapid increases in immigration over the last several decades, including from several countries that have family and gender relationships quite different from those of native-born Swedes. The increase in immigration has greatly diversified Sweden's population. Sweden's experience is in many ways similar to that of many other countries of Europe. However, more than most countries, Sweden encourages immigrant incorporation, emphasizing gender equality in the home and at work via extensive support policies. The financial incentives from the state provide a strong incentive for members of immigrant sub-communities to become "Swedish" in their family formation behavior, pressures that are likely to be most strongly felt by the second generation, who have been socialized in Sweden. However, Swedish family formation patterns challenge the preferences of their parents, which reinforce their origin culture of gender segregation. The preferences they form and the decisions they make will provide important information about the incorporation trajectories likely to be followed by the new immigration to Sweden. Immigrants from Turkey and from Poland represent very different immigrant streams to Sweden. Their different characteristics are likely to shape their relative integration into Swedish society and, in particular, frame the families that they form. Those from Poland are primarily from urban areas, with relatively high levels of education (higher than native Swedes) for both men and women. The level of intermarriage with Swedes is high, even in the first generation, so that Swedish was the only language spoken in most of the homes of Polish immigrants. Nevertheless, cohabitation was rare in the Polish communities from which they came. In contrast to the Polish immigrant pattern, one in three Turkish immigrants came from the countryside. Turkish immigrants had low levels of education, particularly among women. Most Turks were married to other Turks and almost none of the immigrant generation cohabited. It is against this background that we examine the young adult children of these two immigrant groups, focusing on their attitudes and actual choices with regard to forming a partnership via cohabitation or marriage, relative to young adults of Swedish origins (the native born of native parents). We have begun to analyze these questions using logistic regression. For the analysis of restrictive attitudes, we analyze a two-category variable: placing some restrictions on the appropriateness of cohabitation (such as when there are children) or placing no restrictions (as is the case for the vast majority of Swedes). For the analysis of actual cohabitation, we use multinomial logistic regression based on three categories: first union a cohabitation, a marriage, and no coresidential union. The results are complex. There are more differences between groups in attitudes toward cohabitation than in actual partnership formation, although those of Turkish background are both more likely to hold restrictive attitudes and to avoid cohabitation. There is no difference between those of Polish and Swedish background in cohabitation attitudes, but Poles are significantly less likely to actually cohabit, suggesting that either their parents' or partners' preferences are also a factor. We also find, and will test for significance, substantial differences in the factors that predict these outcomes. The results of our analysis will be interpreted in the context of the backgrounds of the immigrant, family policies in Sweden and the patterns of ethnic integration in Sweden.