A conceptual model to address threats to validity in controlled experiments
Amadeu Anderlin Neto, Tayana Uchôa Conte · 2013
Context: During the planning phase of an experiment, the threats to validity must be identified in order to assess their impact over the data. In addition, the actions to address these threats must be defined (if possible). Objective: This paper proposes a conceptual model which contains the relationships between threats to validity and actions to address them. Method: A Systematic Literature Review was conducted to collect data for building the conceptual model. We identified 166 papers published in nine journals and four conferences that reported threats to validity in controlled experiments. Results: We identified 39 threats to internal validity (95 actions to address them), 9 to external validity (23 actions to address them), 10 to construct validity (21 actions to address them), and 8 to conclusion validity (19 actions to address them). Conclusion: By presenting this conceptual model, we intend to assist novice researchers in identifying and addressing the threats to validity of empirical studies.