Dataset and R Code for Correlates of Contestation Against Religious Authorities in Medieval Inquisitorial Testimonies

This deposit contains the dataset and R code necessary to reproduce all statistical analyses reported in the article titled Correlates of Contestation Against Religious Authorities in Medieval Inquisitorial Testimonies. The study examines 4,433 testimonies drawn from 20 inquisition registers across Latin Christendom (1243–1522), investigating how explicit contestations of religious authority correlate with dissident religious affiliation, gender, urban context, and over time. (This is the third version, which incorporates all revisions requested by peer reviewers in the first and in the second round of reviews.) analyses.RComplete R script implementing the analysis workflow: descriptive statistics, register-level beta regressions, testimony-level multilevel logistic regression, and random-forest permutation importance with bootstrap. testimonies.tsvTab-separated, UTF-8 encoded file containing the testimony-level dataset (n = 4,433). Each row corresponds to one testimony. Column documentation:d1_fname – Unique identifier of the document (string).d2_ter – Register short name (string; 20 registers in corpus).d3_title – First 20 characters of the text (for reproducibility checks, due to licence only first 20 characters is allowed).Y – Dependent variable: testimony contains contestation against religious authorities (1 = yes, 0 = no).x1_urban – Urban setting of the register (1 = urban, 0 = non-urban).x2_date – Mean date of the register (decimal year; .5 = 1 July).x3_deprel – Religious affiliation of the deponent (categorical: Apostles, Beguins, Cathars, Guglielmites, Lollards, Waldensians, Other_heterodox, Non_heterodox).x4_gender – Gender of the deponent (1 = male, 0 = female).c1_tok – Number of tokens (linguistic words) in the testimony (integer).c2_lex – Lexical diversity of the testimony (MTLD measure; float).c3_qf – Number of inquisitorial questions in the testimony (integer). testimonies_variables.tsvTab-separated, UTF-8 encoded file, lookup table linking column names of testimonies.tsv to labels (used in tables and figures). results.rdsOutput RDS object (produced by running the R code) containing testimonies, variables, fitted models, marginal effects, and bootstrap variable importance. (Given stochastic nature or ranger-bootstrap, minor difference are to be expected.) Figure1.png300dpi png file of Figure 1 (generated by analyses.R) Figure2.png300dpi png file of Figure 2 (generated by analyses.R)本数据集存档包含复现论文《中世纪宗教裁判证词中针对宗教权威的质疑关联因素》中所有统计分析所需的数据集与R代码。本研究分析了取自拉丁基督教世界20份宗教裁判档案（1243年至1522年）的4433份证词，旨在探究针对宗教权威的明确质疑与异教宗教派系、性别、城市环境以及时间维度之间的关联。（本版本为第三版，整合了一审与二审同行评审专家提出的全部修订意见。） alyses.R：完整的R脚本，实现了整套分析流程：描述性统计、档案层面的β回归、证词层面的多层逻辑回归，以及带自助法（bootstrap）的随机森林置换重要性分析。 testimonies.tsv：制表符分隔、UTF-8编码的证词层面数据集文件（样本量n=4433），每行对应一份证词。 字段说明： d1_fname – 文档唯一标识符（字符串类型） d2_ter – 档案简称（字符串类型，本语料库共包含20份档案） d3_title – 文本前20个字符（用于可复现性校验，因授权限制仅允许使用前20个字符） Y – 因变量：证词中包含针对宗教权威的质疑（1=是，0=否） x1_urban – 档案所属的城市环境（1=城市，0=非城市） x2_date – 档案的平均年份（十进制年份；0.5代表7月1日） x3_deprel – 作证者的宗教派系（分类变量：使徒派、贝居因派、清洁派、古列尔米特派、罗拉德派、韦尔多派、其他异教派系、非异教派系） x4_gender – 作证者的性别（1=男性，0=女性） c1_tok – 证词中的Token（语言学中指单词）数量（整数类型） c2_lex – 证词的词汇多样性（采用MTLD度量法；浮点类型） c3_qf – 证词中的宗教裁判提问数量（整数类型） testimonies_variables.tsv：制表符分隔、UTF-8编码的查找表文件，用于将testimonies.tsv的列名映射为表格与图表制作所用的标签。 results.rds：通过运行R代码生成的RDS格式输出对象，包含证词数据、变量表、拟合模型、边际效应以及自助法（bootstrap）变量重要性结果。（由于ranger-bootstrap的随机性，结果可能存在细微差异。） Figure1.png：由analyses.R生成的300dpi PNG格式图1文件。 Figure2.png：由analyses.R生成的300dpi PNG格式图2文件。