Learning Preferences Survey Dataset – La Paz Campus (2025)
Description
This repository provides an anonymized bilingual survey dataset collected from 144 undergraduate students at the Universidad Nacional de Colombia, La Paz campus, between May and July 2025. The dataset includes academic, demographic, socioeconomic, geographic, inclusion, language, employment, and family-care variables, together with 80 Likert-form items adapted from the Honey-Alonso Learning Styles Questionnaire (CHAEA). Spanish- and English-language versions are supplied, each accompanied by a detailed variable dictionary. Exact response timestamps and direct identifiers were removed and replaced with generated participant codes. The repository also includes an R script for data-quality checks, CHAEA-derived scoring, descriptive statistics, reliability analysis, nonparametric comparisons, sensitivity analysis, exploratory demographic tests, and publication-quality figures. The data are intended to support reproducible research in higher education, educational measurement, survey analysis, and group-level learning-preference profiling.
Files
Steps to reproduce
1. Download and unzip the repository files into a single local folder. 2. Open R or RStudio and set the working directory to the folder containing learning_preferences_analysis.R and learning_preferences_survey_en.xlsx. 3. Install the required R packages, if necessary: install.packages(c("readxl", "dplyr", "tidyr", "stringr", "purrr","ggplot2", "forcats", "scales", "jsonlite")) 4. Run the analysis from the R console: source("learning_preferences_analysis.R") 5. By default, the script reads learning_preferences_survey_en.xlsx. The Spanish-language dataset can be analyzed by providing its filename as an argument: Rscript learning_preferences_analysis.R "learning_preferences_survey_es.xlsx" 6. The script checks the dataset structure, scores the 80 CHAEA-derived items, calculates the Active, Reflective, Theoretical, and Pragmatic profiles, and creates the Analytic and Action-oriented composite scores. It then performs descriptive statistics, reliability analysis, the Friedman test, Kendall’s coefficient of concordance, paired Wilcoxon tests, Holm-adjusted comparisons, effect-size estimation, sensitivity analysis, and exploratory demographic comparisons. 7. Aggregated statistical tables and the analysis summary are saved in the analysis_outputs folder. Publication-ready figures are exported in PNG format at 800 dpi and in vector PDF format to the figures folder.