Research Data and Analytical Code — Road Drainage Infrastructure Diagnostics and Deficiency Indexing in ENSO-Vulnerable Andean Corridors: A STEM-PjBL Field Assessment

Dataset and Analytical Code for: Road Drainage Infrastructure Diagnostics and Deficiency Indexing in ENSO-Vulnerable Andean Corridors: A STEM-PjBL Field AssessmentThis repository contains the complete research data and analytical code supporting the above manuscript, submitted to Sustainability (MDPI, Special Issue: Resilient Infrastructure for Climate Action: The Nexus of Technology, Management, and Innovation). The study presents a STEM-integrated Project-Based Learning (PjBL) diagnostic framework applied to 42 road segments across six corridors in Loja Province, southern Ecuador, where El Niño–Southern Oscillation (ENSO)-driven precipitation extremes pose recurrent threats to road drainage infrastructure. Using ArcGIS Survey123, standardised field data were collected on the presence, cross-sectional geometry, structural condition, and failure modes of four drainage structure typologies: crown gutters, road gutters, hydraulic chutes, and culverts. A Composite Drainage Deficiency Index (DDI; range 0–100) was derived from five equally weighted binary indicators to provide an operationally deployable, georeferenced prioritisation tool for highway maintenance authorities. Statistical analysis was performed in R v 4.5.2 using a fully non-parametric pipeline comprising Shapiro–Wilk normality assessment, Wilson 95% confidence intervals, Fisher's exact tests, Cramér's V with bootstrap confidence intervals, Spearman rank-order correlations with Benjamini–Hochberg false discovery rate correction, Kruskal–Wallis H-tests with ε² effect sizes, Ward.D2 hierarchical clustering with silhouette optimisation, and jackknife leave-one-out sensitivity analysis. This repository includes the raw field dataset (datos_base.xlsx), the complete annotated R script (codigo_R.R), and the full R console output (Consol_R.txt), enabling end-to-end reproducibility of all reported results, figures, and tables from the primary data.数据集与分析代码：易受厄尔尼诺-南方涛动（ENSO）影响的安第斯走廊道路排水基础设施诊断与缺陷指数：一项科学、技术、工程与数学（STEM）项目式学习（PjBL）实地评估。 本仓库包含支撑上述投稿论文的完整研究数据与分析代码，该论文已提交至《Sustainability》（MDPI出版社，特刊主题：气候行动韧性基础设施：技术、管理与创新的交叉融合）。 本研究提出了一套融合STEM的PjBL诊断框架，并将其应用于厄瓜多尔南部洛哈省六条走廊的42个路段。该区域受ENSO驱动的极端降水事件频发，对道路排水基础设施构成持续威胁。 研究采用ArcGIS Survey123工具，针对檐口排水沟、道路边沟、水力滑槽与涵洞四类排水构筑物，收集了其存在情况、横断面几何形态、结构状况与破坏模式的标准化实地数据。 本研究从五个权重均等的二元指标中推导得到综合排水缺陷指数（DDI，取值范围0~100），可为公路养护部门提供可落地部署的地理参考优先级划分工具。 统计分析在R语言4.5.2版本中完成，采用完整的非参数分析流程，涵盖夏皮罗-威尔克（Shapiro–Wilk）正态性检验、Wilson 95%置信区间、Fisher精确检验、带自助法（bootstrap）置信区间的Cramér's V分析、经本杰明尼-霍赫伯格（Benjamini–Hochberg）错误发现率校正的Spearman秩相关分析、带ε²效应量的Kruskal-Wallis H检验、带轮廓系数优化的Ward.D2层次聚类，以及刀切法留一法敏感性分析。 本仓库包含原始实地数据集（datos_base.xlsx）、完整带注释的R脚本（codigo_R.R）以及完整的R控制台输出文件（Consol_R.txt），可实现基于原始数据的所有报告结果、图表与表格的端到端可复现。