Accuracy, efficiency, productivity and researchers’ satisfaction in digital humanities data analysis: Experiment design

Abstract

Data analysis represents the most important group of tasks carried out in research contexts. Due to the current lack of empirical studies about data analysis performance in digital humanities research contexts, we conducted an empirical experiment comparing data analysis performance employed traditional software versus data analysis performance employed software-assistance tools which incorporate cognitive processes in their design. The experiment is designed in terms of accuracy, efficiency, productivity and user satisfaction during the data analysis made by researchers in digital humanities. This documental appendix presents all the materials created and employed for the empirical experiment presented, including the original forms to obtain information regarding the personal and professional profile of the subjects, as well as data analysis tasks defined in two different case studies in digital humanities that allowed us to measure 4 metrics: accuracy (in terms of correctness of the subjects' responses), efficiency (in terms of the time employed by subjects in performed each task), productivity (as a accuracy/efficiency ratio) and researchers’ satisfaction ( The instrument employed is a questionnaire based on Moody’s framework, which evaluates satisfaction based on three concepts: Perceived Usefulness (PU), Perceived Ease of Use (PEOU) and Intention to Use (ITU) in a 5-point Likert scale). The materials can be easily reusable or take as a basis for similar experiments about data analysis performance in digital humanities research contexts.