Data Analysis & Visualization in R

Overview

Coursework for BIO 479/579 (Data Analysis & Visualization in R, Fall 2024) exploring Instacart's online-grocery datasets with the tidyverse. One deliverable analyzes the products catalog (department counts, aisle density, top items, wide-to-long reshaping) and presents it as a slide deck; another is an R Markdown report profiling the orders dataset with summary statistics, tidyverse manipulations, and three ggplot2 visualizations. A separate assignment critiques effective versus ineffective data graphics. **Highlight:** 3,421,083 orders profiled (7 columns)

Key Achievements

• Wrote an R script (BIO579FINAL.R) that summarizes the products catalog with dplyr and visualizes it via ggplot2: department-count bar chart, aisle-by-department density heatmap, top-10 most common products, and a wide-to-long reshape with reshape2/tidyr
• Authored an R Markdown report (MidCourseSummaryAssignmentElsaady.Rmd, knit to HTML) profiling the orders dataset, including base-R and dplyr summary statistics and a unique-user / average-order-interval breakdown
• Performed two tidyverse manipulations on the orders data (weekend-only filtering and group-by day-of-week summaries) and produced three ggplot2 charts: a scatter plot, a bar plot of orders by weekday, and a density plot of days since prior order
• Built a slide presentation (BIO579_Presentation-1.pdf) translating the products analysis into narrative insights with embedded R code for each visualization
• Completed a data-visualization critique assignment evaluating a strong ridgeline plot and a cluttered scatter plot against clarity, color, and annotation criteria

Approach

Datasets are loaded with readr, then inspected for structure and format (str/glimpse) before analysis. Aggregation and filtering use dplyr group-by/summarise pipelines and tidyr/reshape2 for reshaping between wide and long formats. Each visualization uses a distinct ggplot2 geom (bar, tile/heatmap, scatter, density) with theme_minimal styling, and findings are written up alongside the code in R Markdown and presentation form.

Tools & Technologies

Gallery

Results

The orders dataset is profiled as 3,421,083 rows across 7 columns; the products analysis reports departments 13 and 19 as the largest by product count and aisles 33 and 41 as the most diverse. See docs/BIO579_Presentation-1.pdf and docs/MidCourseSummaryAssignmentElsaady.html for the full deliverables.