The use of the names and surnames of individuals as a selection criterion to construct the sampling frame of an investigation is an important strategy in the study of migration. Whether for a descriptive purpose or as a preliminary step in the design of a sample, it allows identifying a subpopulation that would otherwise remain difficult to access. This does not prevent obvious difficulties, typical of data collection methods in general and of this particular application (Alaminos, 1989; Frances et al, 2015; Santacreu, 2016). In this sense, as a strategy to investigate the reality of intra-European migrants, it has been frequently used by the OBETS research group in many national and international research projects, using as reference information telephone directories or, for example, electoral lists with the candidates in local elections. These pages examine their application conditions, as well as evaluate their internal and external validity. The study Political participation as candidates for European residents in Spain, Ministry of Economy and Competitiveness (CSO2012-32930), is taken as a reference case. For the internal evaluation, the information of cases extracted by means of inclusion and exclusion algorithm is reviewed on the basis of onomastic coincidences. The validation review highlights problems related to compound names or second generation. External validation is performed by comparing the results of internal validation with official and documentary information published by the media or political parties. Given the scarce public systematization of the data, this comparison is made in a specific way for different moments of time. The external validation designed corresponds to the logic of the hybrid data, proposed by Euroestat, using multiple sources of information.
Read full abstract