The aim of this study was (1) to examine whether Turkish older migrants are indeed—as is often claimed without solid scientific evidence—lonelier than their peers with no migration background and (2) to determine the factors that account for the differences in loneliness between them. We analysed data of adults aged 50–79 from the first wave of the German Generations and Gender Survey and a supplementary survey of Turkish nationals in Germany (N = 3,248 born in Germany and N = 494 born in Turkey). Differences in degree of loneliness between Turkish and native-born older adults were determined by the six-item Loneliness Scale of de Jong Gierveld. To identify the specific factors contributing to these loneliness differences, a series of multivariate regression analyses were conducted, examining the impact of two groups of risk factors (poor health and low socioeconomic status) and two groups of protective factors (social embeddedness in the family and informal support exchanges) on loneliness. Results showed that feelings of loneliness are indeed more prevalent among older adults of Turkish origin than their German counterparts, which is entirely attributable to their lower socioeconomic status and poorer health. Living with a partner or children, frequent contacts with non-coresident children, emotional support exchange and looking after grandchildren—though important factors to prevent loneliness at the individual level—did not specifically protect Turkish older adults from loneliness, or did so rarely. These findings not only indicate new and challenging directions for further research but also raise questions about the effectiveness of the most common loneliness interventions, which focus on improving number and quality of social relationships.
Nowadays demographers and other social scientists discuss the sharp differentiation between regions in Europe: for example, Reher (1998) contrasted Southern Europe where the family and family ties are ‘strong’ with Western and Northern Europe, where families and family ties are relatively ‘weak’. They base their theoretical ideas on differences in family attitudes and on varying percentages of older adults either living independently or co-residing with adult children. In this paper we address the empirical basis of differentiations between European regions as far as the familial position of older adults is concerned and the outcomes of family embedment for loneliness among older adults. Data come from the UN Generations and Gender Surveys, and encompass the Netherlands, France, Germany, Bulgaria, Russia and Georgia; sample sizes vary between 8.161 and 12.828. Main variables investigated are: filial attitudes, differences in the realization of co-residence or living alone, the frequency of contacts with children, and the quality of relationships with the children. Moreover, a typology of country level differences in familial support arrangements, based on the outcomes of Latent Class Analyses, are taken into account. Outcomes of multivariate analyses show that the countries of Central and Eastern Europe differ significantly from the countries in Western Europe in types of living arrangements, familial attitudes, but also in country level familial support arrangements. Variations between countries and within countries have to be taken into account in explaining the outcomes of family support for social embedment and loneliness of older adults.