Harvard CS50's Introduction to R Programming: Full University Course

Course Overview and R's Strengths

• 💡 This course introduces programming using R, a language popular for statistical computing and graphics in data science.
• 🚀 Learners will progress from basic R usage to packaging, testing, and sharing R code.
• 🎯 R is highlighted as a language built for data analysis, making it ideal for fields like data science, visualization, research, and statistics.

Setting Up and First Program

• 💻 RStudio is introduced as the integrated development environment (IDE) specifically designed for R, featuring a console for line-by-line execution and a file editor for full programs.
• 📝 The first R program, "Hello, World!", is created by defining a print() function within an R file, demonstrating basic syntax and execution via the "Run" button.
• 🐞 Debugging is introduced as the process of finding and fixing errors, exemplified by intentionally mistyping the print function to trigger an error and then correcting it.

User Input and Dynamic Output

• 💬 The readline() function is used to capture user input, prompting the user for their name.
• 🤝 String concatenation is explained using the paste() function to combine literal strings with user input, creating dynamic greetings like "Hello, Carter".
• ⚙️ The paste0() function is introduced as a more concise alternative to paste() for concatenating strings without default separators, and cat() is mentioned as a function that concatenates strings and prints them to the console.

Variables, Data Types, and Arithmetic

• 📦 Variables (objects) are used to store data, with name storing user input and greeting storing the combined string.
• 🔢 R supports basic arithmetic operators: addition (+), subtraction (-), multiplication (*), and division (/).
• 🧱 Data types (storage modes) like character strings, doubles (decimal numbers), and integers (whole numbers) are discussed, along with coercion using functions like as.integer() to convert data types.

Working with Data Frames and Files

• 📂 R Studio's environment pane displays stored objects, and functions like ls() (list) and rm() (remove) manage these objects.
• 📄 CSV files (comma-separated values) are introduced as a common format for storing tabular data.
• 📊 The read.csv() function is used to import CSV data into R, creating data frames.
• 🔍 Data frames can be accessed using bracket notation (e.g., votes[1, 2]) or more robustly using dollar sign notation (e.g., votes$poll) to access columns by name.
• 🧮 Vectorized operations are highlighted, where functions like sum() can operate on entire vectors efficiently, and vector arithmetic allows element-wise operations (e.g., votes$poll + votes$mail).

Advanced Data Handling and Visualization

• 🌐 Online data sets can be read directly into R using functions like read.csv() with a URL.
• 📊 Functions like n_row() and n_col() provide dimensions of data frames.
• ❓ Unique values within a column can be found using the unique() function.
• 🏷️ Factors are introduced for representing categorical data, allowing labels to be assigned to numerical codes (e.g., 1='Yes', 2='No').
• 🧩 Data tidying principles are explained: each observation is a row, each variable is a column, and each cell is a single value.
• ↔️ pivot_wider() reshapes data from long to wide format, turning row values into column headers.
• 📈 ggplot2 is introduced for data visualization, using layers (geoms, scales, labels, themes) to build plots like bar charts (geom_col), scatter plots (geom_point), and line graphs (geom_line).

Programming Constructs and Best Practices

• 🔄 Loops (repeat, while, for) enable code repetition for tasks like repeatedly prompting for valid user input.
• 🛠️ Functions (function()) allow code modularization, reusability, and parameterization (e.g., get_votes() with a prompt parameter).
• ⚠️ Error handling is crucial, using is.numeric(), is.na(), warning(), stop(), and suppressWarnings() to manage invalid input and unexpected data.
• 🧪 Unit testing with the testthat package ensures functions behave as expected, using expect_equal(), expect_warning(), and expect_error() for various scenarios.
• 📦 Package development involves organizing code into folders (R/, man/, tests/), writing DESCRIPTION and NAMESPACE files, and using devtools functions (load_all(), build(), use_testthat(), use_r()) to manage and build packages for sharing.
• ✍️ Documentation is created using R's markup language (.Rd files) within the man/ folder, explaining function usage, parameters, and examples.