The effect of reasoning strategies on success in early learning of programming: lessons learned from an external experiment replication

摘要: Background. Literal or theoretical replications are important to evaluate and assess empirical results. However, there still few in software engineering, fewer external replications, i.e., developed by researchers other than the original ones. Aim. This paper discusses difficulties found lessons learned from performing two literal of an experiment involving human subjects. Results. Our results apparently contradict conclusions experiment. several differences context made it difficult achieve valid comparability. Conclusion. Experiments subjects should collect report as many qualitative information possible, so can be related conditions under which hypothesis were true. Besides, given this study, replication does not seem best strategy for experiments engineering.