The Cost of Being “Late”: Tracing Misalignment and Misdiagnosis Through Relational Architectures

Simple item page
dc.contributor.advisorCandia Vallejos, Cristian
dc.contributor.advisorSoto-Icaza, Patricia
dc.contributor.authorDavyt Colo, Joselina Beatriz
dc.coverage.spatialSantiago
dc.date.accessioned2026-05-26T16:22:37Z
dc.date.available2026-05-26T16:22:37Z
dc.date.issued2026
dc.descriptionDissertation submitted to the Facultad de Gobierno of the Universidad del Desarrollo for the degree of Doctor in Social Complexity Sciences
dc.description.abstractThe main focus of this thesis is the investigation of the structural origins of temporal disadvantage, manifested as reduced academic persistence and diagnostic delay, across two distinct domains: higher education and clinical diagnosis. Adopting a Computational Social Science framework, we employ network analysis not merely as a tool, but as a theoretical lens to visualize and quantify the relational architectures that generate structural misalignment between individuals and institutions. The first study examines informational inequality in higher education. Using administrative data from 1.6 million applicants in Chile and replicating findings in Portugal, we construct a network of degree preferences to quantify Preference Misalignment—the distance between a student’s true interests and their enrolled program. We demonstrate that this topological distance is a robust predictor of first-year retention: students with misaligned preferences face a significantly higher risk of persistency, a penalty that remains even among high-performing students. The second study addresses recognitional inequality in autism. Analyzing a clinical sample of autistic children without intellectual disability, we integrate bipartite and multilayer networks to decode the mechanisms behind the diagnostic delay in autistic girls (averaging two years later than autistic boys). We identify while structural connectivity (i.e., high betweenness and participation coefficient) accelerates diagnosis in boys, it significantly delays it in girls. Conversely, phenotypic entropy (systemic disorder) facilitates recognition in females. Multilayer analysis further revealed that autistic girls exhibit a differentiated physiological and cognitive architecture, in which high cognitive ability is structurally coupled with increased psychological burden. Together, these findings reveal that being "late" is rarely an individual accident, but a consequence of institutional architectures that render certain profiles illegible. By mapping these invisible structures, this work provides empirical evidence for policy interventions aimed at reducing the structural friction that generates inequality.
dc.format.extent89 p.
dc.identifier.doihttps://doi.org/10.52611/11447/10774
dc.identifier.urihttps://hdl.handle.net/11447/10774
dc.language.isoen
dc.publisherUniversidad del Desarrollo. Facultad de Gobierno
dc.subjectComputational Social Science
dc.subjectNetwork Science
dc.subjectInstitutional Misalignment
dc.subjectAutism & Sex Differences
dc.subjectBehavioral Data Science
dc.titleThe Cost of Being “Late”: Tracing Misalignment and Misdiagnosis Through Relational Architectures
dc.typeThesis
dcterms.accessRightsAcceso abierto

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
The Cost of Being “Late”: Tracing Misalignment and Misdiagnosis Through Relational Architectures.pdf
Size:
3.47 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
347 B
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Tesis Doctorales

Implementado por OpenGeek Services