LAUSR

This paper explores immigrants’ transnational mobility in Germany. It uses data of the Socio-Economic Panel Study and four indicators regarding frequency, length, and total duration of visits to the country of origin. The study applies factor mixture analyses in order to investigate whether (a) the observed indicators refer to a uniform underlying construct of transnational mobility and (b) the relationship between the latent construct and the observed indicators establishes in a uniform manner for all respondents. The most reliable model distinguishes three latent classes of immigrants, thus indicating no uniform underlying construct of transnational mobility. Theoretically consistent findings could be derived for about 58 % of the 4019 respondents. However, the relation between the observed indicators and the latent variable diverges substantially for the remaining 42 %. Thus, the findings indicate that the commonly applied indicator of return visits largely fails to assess transnational mobility. Rather, different groups of immigrants engage very diversely in visits to the country of origin. The findings stimulate a variety of conceptual problems future theoretical and empirical research needs to tackle.