作者: Rafael Maiani de Mello , Pedro Correa da Silva , Per Runeson , Guilherme Horta Travassos
关键词: Conceptual framework 、 Software engineering 、 Sampling frame 、 Sampling (statistics) 、 Experimental software engineering 、 Computer science 、 The Conceptual Framework 、 Systematic sampling 、 Population 、 Data science 、 Crowdsourcing
摘要: Context: The low quality and small size of samples in empirical studies software engineering hamper the interpretation generalization their results. Therefore, enlarging sample sizes improving represent an important research challenge. Goal: We aim to define a conceptual framework, including requirements for establishing adequate sources sampling subjects surveys. Method: use previous experience on applying systematic strategies combined with contemporary web technologies previously executed surveys, organize framework. analyze its application different sampling. Results: framework was observed be feasible after nine large-scale Conclusions: analyzed crowdsourcing tools do not support essential considered sampling, while free-lancing professional social network do.