Computer Literacy Course LMS Interaction Dataset: Metadata and Data Dictionary (Restricted Access)

This record provides the metadata and data dictionary for the anonymized Moodle Learning Management System (LMS) interaction dataset used in the study "From LMS Logs to Pedagogical Insights: An Explainable Machine Learning Approach for Early Detection of At-Risk Students in Higher Education." DATASET OVERVIEW:- 2,618 anonymized student records (cleaned from an initial pool of 3,021)- 32 course sections, fully online Computer Literacy course- Semester: 2021-2022 Fall- Platform: Moodle LMS- 25 engineered features across three groups: 9 content-type interaction counts, 12 weekly module sums, and 4 temporal engagement ratios- Binary outcome variable: Pass (n=1,463, 55.9%) / Fail (n=1,155, 44.1%) DATA ACCESS RESTRICTION:The full dataset is NOT publicly available due to student privacy regulations, specifically the Turkish Personal Data Protection Law (PDPL, Law No. 6698) and the European Union General Data Protection Regulation (GDPR). Although student identities were anonymized using the SHA-256 hash algorithm, the combination of behavioral features across 32 sections could potentially enable re-identification. The study was approved by the Başkent University Social and Humanities Sciences and Arts Ethics Committee (Decision No: 17162298.600-31), and the terms of ethical approval prohibit unrestricted public sharing of the underlying data. HOW TO REQUEST ACCESS:Researchers who wish to access the full anonymized dataset for replication or further analysis may contact the corresponding author: Habibe Karayiğit E-mail: [e-mail] Access will be granted on a case-by-case basis, subject to:(1) Institutional data governance approval(2) Execution of a data sharing agreement ensuring PDPL/GDPR compliance(3) Confirmation that the data will be used solely for non-commercial research purposes FILES INCLUDED IN THIS RECORD:- README.md: Overview of the dataset, variables, and access procedures- data_dictionary.csv: Complete descriptions of all 25 features and the target variable