Analytics
Student Learning Preference Analysis
Survey-based analysis of traditional and AI-enabled study tools
Overview
A survey research project under Professor Labadi analyzing how students use traditional resources, online platforms, and LLM tools for learning.
Problem
Students now mix traditional resources with online platforms and AI assistants. The project investigates which tools students prefer and whether demographic factors meaningfully affect outcomes.
Dataset
A Google Forms survey reduced from 100 initial responses to 60 balanced samples through stratified sampling and integrity checks.
Approach
Performed Excel-to-R data cleaning, EDA, regression analysis, ANOVA, and sampling checks to evaluate study efficiency and preference differences.
Results
Estimated LLM impact on study efficiency with p-values above 0.05, while documenting preference patterns and limitations from the sample design.
Lessons Learned
Survey work needs careful sampling, clear variable definitions, and honest limits around sample size. Good communication made the statistical findings easier to act on.
Model / Pipeline
The implementation combines R, Regression Analysis, ANOVA, Excel, Google Forms, EDA with a repeatable workflow for data preparation, evaluation, and communication.