Informer Availability Bias: When Data Fidelity and Reach Move in Opposite Directions
Informing Science: The International Journal of an Emerging Transdiscipline · 2026 · 29 · 19
DOI: 10.28945/5898 ↗Kurz gefasst
Fachartikel in der Zeitschrift Informing Science: The International Journal of an Emerging Transdiscipline, 2026. Verfasst von Sira Satayapaisal und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Aim/Purpose This paper examines what happens to an informing system when the people it exists to protect are the least likely to supply it with information. It also asks why the system’s own quality checks cannot detect this. The problem is illustrated with school screening, where a questionnaire is used to find students drifting toward isolation, absenteeism, conflict, or distress.
Background
Informing science views informing as a system comprising an informer, a delivery system, and a client. Earlier work showed that the client had been neglected. We argue that the informer has also been taken for granted. In many real systems, a person must first agree to become an informer, and this willingness is not evenly distributed. When it is uneven, the people a system is meant to help may be the ones it cannot see. Methodology We propose informer availability bias, separate it from the classical taxonomy of missing data, and develop a three-step diagnostic to detect it. The diagnostic is applied to an open dataset of 110 students from a secondary school in Spain. Only 86 answered the questionnaire. The other 24 appear because peers named them. A simulation calibrated to this school compares random and biased non-response at rates of 5% to 40%, with 2,000 replications per rate. Contribution The paper names and defines informer availability bias and offers a simple framework. It also provides a three-step diagnostic that any holder of relational data can use, with practical steps for system designers. The simulation shows that, under biased non-response, data fidelity and system reach move in opposite directions.
Findings
Outgoing ties of non-respondents in the dataset were reconstructed, not observed. In reported ties, non-respondents received 42% of the nominations expected under an even spread. Deleting them changed which students appeared most isolated. At 20% simulated non-response, biased non-response achieved a reach of 43% compared to 56% for random non-response, but had higher rank fidelity (0.972 vs 0.957). Recommendations for Practitioners Keep records of people who did not answer rather than deleting them. Check whether another source, such as peer reports or school records, can show where they stand. Every report should state how many people are missing and how they differ from respondents. This applies to any system based on voluntary self-report. Recommendations for Researchers High agreement between observed and true data does not mean that a system works for the people it is meant to serve. Report reach into the target group as a separate figure. In studies of relational data, say whether non-respondents were kept and whether any of their data were reconstructed. Impact on Society Schools and other services often use screening to decide who gets help first. If the people who most need help are also the least likely to respond, they can be missed, and the system’s own checks will not show it. Reporting reach makes this problem visible. Future Research The proposition should be tested where a non-voluntary second channel exists, such as attendance or health records. Replication in other schools and outside education is needed. Networks with a weaker link between in-degree and out-degree (0.75 here) may show a smaller divergence. Methods to estimate reach without a second channel are also needed.
Zitieren
APA
Satayapaisal, S., Akrajindanon, R., Netinant, P., & Tantiathimongkhon, T. (2026). Informer Availability Bias: When Data Fidelity and Reach Move in Opposite Directions. Informing Science: The International Journal of an Emerging Transdiscipline, 29, 19. https://doi.org/10.28945/5898
Vancouver
Satayapaisal S, Akrajindanon R, Netinant P, Tantiathimongkhon T. Informer Availability Bias: When Data Fidelity and Reach Move in Opposite Directions. Informing Science: The International Journal of an Emerging Transdiscipline. 2026;29:19. doi:10.28945/5898
BibTeX
@article{Satayapaisal2026,
author = {Satayapaisal, Sira and Akrajindanon, Ravipa and Netinant, Patamaporn and Tantiathimongkhon, Theerawut},
title = {Informer Availability Bias: When Data Fidelity and Reach Move in Opposite Directions},
journal = {Informing Science: The International Journal of an Emerging Transdiscipline},
year = {2026},
volume = {29},
pages = {19},
doi = {10.28945/5898}
}
EUVIMED