Lecture Notes

Announcements

• Discord Server: An undergraduate learning assistant has set up a Discord server for student use. The lecturer will not use this resource, but students are encouraged to engage.
• Course Project:
  • Project Name: Prison Plot
  • Tasks: Recreate a plot with R code
  • Assumptions: Completion assumes reading of three specific book chapters
  • Due Date: September 27th

R Programming: Vectors and Indexing

Vectors

• Types: Numeric, character strings, logical (true/false)
• Operations: Creation and manipulation using functions

Indexing

• Access Specific Elements: Use square brackets []
  • Example: randnums[5] gives the fifth element
• Access Range of Elements: Use sequence notation
  • Example: randnums[1:3]
• Functions:
  • mean(): Calculate average, e.g., mean(randnums[1:3])
  • length(): Find the number of elements

Accessing Elements

• Last Element: Use length function
  • Example: randnums[length(randnums)]
• Last N Elements: Use sequence and length function
  • Example: randnums[(length(randnums)-4):length(randnums)]

Good vs Bad Practice

• Hardcoding indices is bad practice; using functions like length() is better.

Help Documentation in R

• Access using ?function_name
• Provides detailed information on function arguments
• Example: Understanding mean() function and na.rm to handle missing values

R Basics: Data Frames

Data Frame Creation

• Definition: Equivalent to a spreadsheet in R
• Syntax: data.frame(variable1, variable2, ...)
• Requirements: Equal number of entries in each column

Manipulation

• Single Column Access: dataframe$column_name
• Mean Calculation with NA Handling: mean(dataframe$column, na.rm=TRUE)
• Add New Columns: dataframe$new_column = operation

Subsetting Data

• By Rows: dataframe[row_indices, ]
• By Columns: dataframe[, column_indices]
• By Specific Rows and Columns: dataframe[row_indices, column_indices]

Basic Statistical Concepts

Population and Sample

• Population: Entire collection of observations
• Sample: Subset of the population
• Analogy: Pot of soup (population) and a sip (sample)

Parameters and Statistics

• Parameter: Measurement of a population, denoted by Greek letters (e.g., μ)
• Statistic: Measurement of a sample, denoted by Latin letters (e.g., x̄)

Descriptive vs Inferential Statistics

• Descriptive Statistics: Summarize data (e.g., mean, median)
• Inferential Statistics: Make inferences about populations based on samples

Random Sampling

• Method ensuring every population element has an equal chance of selection
• Important for reducing bias; ideal in survey research

Realistic Research Constraints

• True random sampling is often impractical due to ethical and practical issues
• Non-random samples, like convenient samples, can introduce biases

Conclusion

• Basic R concepts covered are foundational for future topics
• Importance of reading course material for deeper understanding
• Encouragement for questions and engagement in upcoming sessions