EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Data-Driven Analysis of Global Asylum Decisions: Comparative Patterns and Implications for Refugee Protection Decision-Making

Kirti, Dr.Anubha Dwivedi

Computer and Decision Making: An International Journal · 05.11.2026 · 3 (6S) · 2589-2602

DOI: 10.1922/comdem.v3i6s.18997 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 05.11.2026. Verfasst von Kirti und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Asylum decision-making processes around the world are highly geographically, institutionally, and temporally diverse, which poses difficulties in providing consistent refugee protection assessment. This study provides an analysis of asylum decision patterns around the world and assesses a statistical approach based on data for comparative analysis and evidence-based institutional decision-making. The analytical units (7,335) of the secondary data (2000-2025) from the United Nations High Commissioner for Refugees (UNHCR) containing 9,251,218 individual asylum decisions were analysed. To analyze the recognition, rejection and complementary protection outcomes, descriptive statistics and geographical comparisons were used as well as a penalized partially pooled hierarchical model (P1). Baseline comparison, component ablation, rolling temporal validation, cluster-bootstrap analysis, empirical uncertainty calibration, and exploratory statistical monitoring were used in model evaluation. The results showed a overall substantive protection rate of 48.08%, showing significant geographical variation. P1 decreased the log loss by 6.40%, Brier Score by 19.47%, and weighted mean absolute error by 14.29% when compared with baseline B4. Predictive performance was best achieved by origin-country effects. The empirical prediction interval calibration had a 94.87% group-level coverage with the nominal level of 95%. Generalizability and operational reliability were, however, limited due to temporal instability, wide prediction intervals, sensitivity to reporting-completeness, and limited recall in anomaly-detection. The results show how the use of hierarchical modelling can enhance composition-aware comparison of aggregate outcomes of asylum and facilitate transparent institutional assessment. However, such statistics can't make up for procedural fairness, discrimination, or legal compliance. Further temporal independent validation, increased monitoring sensitivity and incorporation of contextual variables is needed before operational implementation.

Zitieren

APA

Kirti, & Dwivedi, D. (2026). Data-Driven Analysis of Global Asylum Decisions: Comparative Patterns and Implications for Refugee Protection Decision-Making. Computer and Decision Making: An International Journal, 3(6S), 2589-2602. https://doi.org/10.1922/comdem.v3i6s.18997

Vancouver

Kirti, Dwivedi D. Data-Driven Analysis of Global Asylum Decisions: Comparative Patterns and Implications for Refugee Protection Decision-Making. Computer and Decision Making: An International Journal. 2026;3(6S):2589-2602. doi:10.1922/comdem.v3i6s.18997

BibTeX
@article{Kirti2026,
  author = {Kirti and Dwivedi, Dr.Anubha},
  title = {Data-Driven Analysis of Global Asylum Decisions: Comparative Patterns and Implications for Refugee Protection Decision-Making},
  journal = {Computer and Decision Making: An International Journal},
  year = {2026},
  volume = {3},
  number = {6S},
  pages = {2589--2602},
  doi = {10.1922/comdem.v3i6s.18997}
}
Kostenlos starten →