Context is Key for Reproducibility of Empirical Studies in Human-Computer Interaction
(2025) ACM Conference on Reproducibility and Replicability (REP ’25) — Location: University of British Columbia, Vancouver, Canada (29.July.2025)
Files
Ampliatum-REP2025.pdf
Preprint
Open Access - Adobe PDF
- 1.56 MB
Details
- Authors
Vatavu, Radu-Daniel 
- Abstract
- Although most reproducibility frameworks are generic enough to apply across a wide range of scientific domains, they may fail to capture the specific dimensions relevant to particular fields, such as Human-Computer Interaction, where interactive prototypes must be evaluated in the context of use involving users, devices, and environments. To address this aspect, we introduce Ampliatum, a contextual reproducibility framework that extends general replication types to cover interactive systems with (1) replication along various contextual dimensions, (2) extension and generalization of previous studies using different populations, and (3) repurposing studies with goals different from those of the original research. To demonstrate the applicability and utility of our framework, we report case studies in gesture-based input. In our first example, we classify existing gesture elicitation studies using our revised terminology. In the second case, we replicate a previous user study under different contextual conditions. Finally, we propose recommendations for the implementation of replicability in Human-Computer Interaction.
- Affiliations
Citations
Vanderdonckt, J., & Vatavu, R.-D. (2025). Context is Key for Reproducibility of Empirical Studies in Human-Computer Interaction. Proceedings of ACM Conference on Reproducibility and Replicability. Published. ACM Conference on Reproducibility and Replicability (REP ’25), University of British Columbia, Vancouver, Canada. https://doi.org/10.1145/3736731.3746142