Data Analysis with R

Topics: introduction, data analysis, statistics, modules

Welcome to the exciting world of data analysis!

This section will help you move beyond basic R programming and show you how to turn health and clinical data into meaningful insights that can inform patient care, research, and medical education.

You’ll learn to use powerful R packages such as:

dplyr — for cleaning and analyzing patient datasets
ggplot2 — for creating clear, evidence-based visualizations
stats (built-in) — for performing statistical tests and clinical comparisons

By the end of this section, you’ll be able to:

Load and explore datasets from sources such as hospital records or surveys
Clean and transform messy data (e.g., missing values in lab tests)
Visualize patterns such as blood pressure trends or lab result distributions
Perform basic statistical analyses to support clinical and research conclusions

Introduction to Data Analysis Packages

Essential packages

Before diving into analysis, let’s explore the core R packages that make data analysis possible — and how they apply to the medical field.

library(dplyr)     # Data manipulation and analysis
library(tidyr)     # Data reshaping
library(ggplot2)   # Data visualization
library(readr)     # Fast CSV reading

There will be times when there are warning messages such as:

Warning message:
package ‘ggplot2’ was built under R version 4.5.3 

and it’s completely fine. The warning message just means that the package in question was created in an R version that is lower than your current version.

These packages are all part of the tidyverse — a coherent collection of R packages designed for data science. You can install them all at once:

install.packages("tidyverse")
library(tidyverse)

1. dplyr — For Data Manipulation

What it does: Handles tabular data (like Excel sheets or patient records) using intuitive, readable verbs.

library(dplyr)

# Example patient dataset
data <- data.frame(
  Name           = c("Ana", "Luis", "Maria"),
  Age            = c(34, 45, 29),
  Blood_Pressure = c(120, 135, 110)
)

# View average blood pressure
mean(data$Blood_Pressure)                               # 121.6667

💡Use case: Compute the average blood pressure of patients in a study.

2. Base R — For Numerical Calculations

What it does: Performs fast mathematical operations on vectors and matrices — no extra package needed.

# Systolic blood pressure readings
bp <- c(120, 130, 110, 125, 140)

# Compute mean and standard deviation
cat("Mean BP:", mean(bp), "\n")           # Mean BP: 125
cat("SD BP:",   sd(bp),   "\n")           # SD BP: 11.18034 

💡Use case: Analyze variability in blood pressure readings from a clinical trial.

3. ggplot2 — For Visualization

What it does: Creates elegant, layered plots and charts.

library(ggplot2)

ages <- c(23, 45, 34, 50, 29, 45, 29, 23)
df_ages <- data.frame(Age = ages)

ggplot(df_ages, aes(x = Age)) +
  geom_histogram(
    bins  = 5,
    fill  = "#2C7FB8",
    color = "white"
  ) +
  labs(
    title = "Age Distribution of Study Participants",
    x     = "Age (Years)",
    y     = "Count"
  ) +
  theme_classic(base_size = 35) +
  theme(
    plot.title = element_text(
      face     = "bold",
      size     = 20,
      hjust    = 0.5
    ),
    axis.title = element_text(
      face     = "bold",
      size     = 20
    ),
    axis.text = element_text(size = 20)
  )

💡 Use case: Visualize the age distribution of your study participants.

4. ggplot2 — For Statistical Data Visualization

What it does: Makes elegant statistical plots with very little code.

library(ggplot2)

df <- data.frame(
  Gender      = c("Male", "Female", "Female", "Male"),
  Cholesterol = c(190, 170, 210, 200)
)

ggplot(df, aes(x = Gender, y = Cholesterol, fill = Gender)) +
  geom_boxplot(
    width         = 0.5,
    alpha         = 0.8,
    linewidth     = 0.7,
    outlier.shape = NA
  ) +
  labs(
    title = "Cholesterol Levels by Gender",
    x     = "Gender",
    y     = "Cholesterol (mg/dL)"
  ) +
  theme_classic(base_size = 35) +
  theme(
    plot.title = element_text(
      face     = "bold",
      size     = 20,
      hjust    = 0.5
    ),
    axis.title = element_text(
      face     = "bold",
      size     = 20
    ),
    axis.text = element_text(
      size    = 20,
      color   = "black"
    ),
    legend.position = "none"
  )

💡 Use case: Compare cholesterol levels across genders or groups.

5. stats — For Statistical Analysis

What it does: Performs statistical tests, correlations, and probability functions. It is built into R — no installation needed.

# Two groups of fasting blood sugar values
group_A <- c(95, 100, 110, 120, 130)
group_B <- c(90,  92,  88,  85,  93)

result  <- t.test(group_A, group_B)
cat("t-statistic:", result$statistic, "\n")     # t-statistic: 3.261195 
cat("p-value:",     result$p.value,   "\n")     # p-value: 0.02700263

💡 Use case: Compare two treatment groups’ fasting blood sugar levels.

In short

R offers powerful tools for handling, visualizing, and analyzing data.

dplyr is your best friend for managing patient records or clinical datasets.
Base R enables fast, vectorized calculations on medical measurements.
ggplot2 helps you visualize trends clearly for research or presentations.
stats (built-in) provides statistical tools to support evidence-based conclusions.

Together, these packages form the foundation of data analytics in R.

Summary Introduction to Data Frames