portfolio project / reproducible regression

Regression Lab in R

Built reproducible R case studies for data cleaning, imputation, transformation testing, and regression modeling using synthetic educational datasets.

01

Goal

Turn two college regression case studies into a cleaner R project that is easy to run, review, and explain.

02

Build

Organized scripts, raw data paths, missing-data handling, transformations, model outputs, and plots into a reproducible workflow.

03

Result

Turned the class work into a clear project archive with scripts, outputs, and notes that make the modeling steps easy to follow.

build notes

Choices and small problems

  • Kept the class data synthetic and labeled clearly so it does not read like real medical or genetic research.
  • Used MICE for missing values instead of throwing away useful rows too quickly.
  • Kept the selected transformations from the original work, then made the outputs easier to rerun and inspect.
  • Saved summaries, ANOVA tables, coefficients, and plots as separate project outputs.

visual notes

Actual regression graphs

These graphs were generated from the uploaded regression data files in the project folder.

Scatterplot of Project 1 Part A IV and DV values with a fitted regression line
Part A: IV/DV scatterplot with fitted regression trend.
Graph selector
Now viewing: Part A regression
02 figs