Investigating bias in the search phase of Software Engineering secondary studies

Context. Researchers are increasingly resorting of secondary studies (e.g. systematic literature reviews and mapping studies) in Software Engineering. This method is strongly dependent on the source of primary studies adopted, which is a bias. We did not find guidelines or benchmarks to evaluate the sources in a systematic way. Objective. In this paper we aim to tackle the selection of electronic data sources while conducting such kind of studies evaluating the equilibrium between the volume and number of relevant papers. Method. In this sense, we proceed towards a secondary study to analyze the overlapping of three different electronic data sources. We also compared our results with other similar studies. Results. Our results show minimum overlapping and no effortless combination of electronic data sources at all. Conclusion. We conclude that researchers shall resort of completeness to work with a feasible set of papers to review. Specially in secondary studies adopting general and no standardized terms