Evidence-Graded Bounds Analysis (EGBA): Protocol for Application
This protocol sets out how to apply Evidence-Graded Bounds Analysis (EGBA) in a study. EGBA compares two decision options when the quantities that decide the comparison are only partly observed and the available evidence differs in quality. The protocol fixes, before any result is computed, the decision function, the source register and its evidence classes, the rules that turn observations into bounds, the evidence ladder, the paired evaluation, the breakdown frontier and the reporting labels. It also states how amendments, related-party sources, transferred values and exploratory runs are handled. The aim is that any researcher can apply EGBA in the same way and that a reader can check every step. The protocol builds on partial identification (Tamer, 2010), breakdown frontiers (Masten & Poirier, 2020) and evidence grading (Balshem et al., 2011). It is a procedural companion to the EGBA methodological note by the same author This protocol sets out how to apply Evidence-Graded Bounds Analysis (EGBA) in a study. EGBA compares two decision options when the quantities that decide the comparison are only partly observed and the available evidence differs in quality. The protocol fixes, before any result is computed, the decision function, the source register and its evidence classes, the rules that turn observations into bounds, the evidence ladder, the paired evaluation, the breakdown frontier and the reporting labels. It also states how amendments, related-party sources, transferred values and exploratory runs are handled. The aim is that any researcher can apply EGBA in the same way and that a reader can check every step. The protocol builds on partial identification (Tamer, 2010), breakdown frontiers (Masten & Poirier, 2020) and evidence grading (Balshem et al., 2011). It is a procedural companion to the EGBA methodological note by the same author
Authors
- Seyed Mehdi Alavi (ORCID: https://orcid.org/0009-0003-8430-211X)
Institutions
- Széchenyi István University (HU)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-13
- DOI
- https://doi.org/10.5281/zenodo.24202017
- Primary Topic
- Advanced Causal Inference Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00