Last saved: 30 August 2002 Quantifying the transnational: Using numbers in research on transnational migration Paper presented at the methodology workshop ‘Possibilities and limitations in transnational migration research’, Department of Anthropology, Gothenburg University, 27 May 2002. Jørgen Carling „ International Peace Research Institute, Oslo (PRIO) „ jorgen@prio.no The bulk of research on transnational migration has been conducted by anthropologists and other social scientists with an orientation towards qualitative research methods. 1 In fact, ethnographic approaches are fundamental to understanding and analyzing the transnational aspects of migra- tion in specific contexts. In this paper, I argue that research on transnational migration can gain from an integration of quantitative methods with ethnographic and other qualitative approaches. While there is clearly a potential for research that is based on advanced quantitative methods, I will concentrate on the use of numerical data by qualitatively oriented researchers who are not trained in statistics. Throughout the paper, I will use examples from my own research on migra- tion from Cape Verde (Carling 1997, 2001, 2002). The paper consists of four parts. First, I discuss different purposes of using quantitative data. Second, possible data sources are considered. Third, I examine the various objects of quantification within transnational migration. Finally, the potentials of using quantitative data are discussed. Purposes of quantification Resulting from the priorities outlined above, thorough quantitative analysis or modelling will only be discussed quite briefly at the end of this section. Instead, I will concentrate on supplementary functions of numerical data. This includes the use of statistics to explore a largely unknown empiri- cal reality, the analysis of qualitative data in light of statistical information, and the study of dis- course that embodies quantitative elements. Exploration Inventive use of statistics has a great potential for exploring an empirical reality that could be spar- ingly documented by social scientific research. This is especially the case in industrialized coun- tries where large amounts of reliable numerical data are collected within a well-organized and easily accessible system of national statistics. While descriptive statistics on the number of immi- 1 Exceptions with extensive use of quantitative data combined with an explicitly transnational paradigm include Brown (1997), Chevannes and Ricketts (1997), Guarnizo (1997) and Massey et al. (1994). Many others have based their research on surveys covering much the same in terms of transnational practices and mobility patterns, but without the vocabulary of transnationalism. A case in point is the rich data collected through the joint Eurostat/NIDI project ‘Push and pull factors of international migration’ European Commission (2000).