Data Types in R

Data Types in R

• Character: Text data (e.g., names, qualitative data).
• Numeric: Real numbers, including integers and decimals (e.g., 1.1, 3.14).
• Integer: Whole numbers (e.g., 0, 1, 2).
• Complex: Numbers with real and imaginary parts.
• Logical: Boolean values, TRUE or FALSE.
• Dates: Represented as date objects (e.g., "2024-06-01").
• Factors: Categorical data stored as integers with labels.

Operators in R

• Logical Operators:
  • ==: Equal to.
  • !=: Not equal to.
  • >=: Greater than or equal to.
  • %in%: Element in a vector.
  • %!in%: Element not in a vector.
• Boolean Values:
  • TRUE and FALSE: Used in logical operations.

Subsetting Data

• Full outer join: The data has all rows in x and all rows in y. Argument: all = TRUE.
• Left outer join: Resulting data has all rows in x. Argument: all.x = TRUE.
• Right outer join: Resulting data has all rows in y. Argument: all.y = TRUE.

Long vs. Wide Format

• Long Format: Multiple observations per ID/grouping variable. Used for repeated measures or longitudinal data.
• Wide Format: Each subject's repeated measures are in a single row with multiple columns.

Data Visualisation

• Scoring:
  • Summing Scores: Add individual item scores directly.
  • Using rowMeans: Calculate the mean of item scores for each participant, excluding missing data.

Cronbach's alpha

• Purpose: Used to assess the internal consistency or reliability of a scale.
• Why We Do Reliability: To ensure that the scale measures consistently across different items.
• Calculation: Collect item scores and use the alpha() function from the psych package.

Data to Ink Ratio

• Definition: The ratio of data presented to the amount of ink used in the graph.
• Goal: Maximize data-to-ink ratio to ensure the graph is efficient and not cluttered.
• Good Practices: Remove unnecessary borders, gridlines, and background colours; use simple, clean designs.

Types of Graphs/Plots

• Histogram: Visualise the distribution of a continuous variable.
• Density: Show the distribution of a continuous variable using a smooth curve.
• Stack dotplot: Display individual data points, useful for small datasets.
• Barchart: Compare counts or portions across categories.
• Scatterplot with regression line: Show the relationship between two continuous variables.
• Violin plots: Combine boxplots and density plot to show data distribution.

GLM Part 1

• Line of Best Fit: The line that best represents the data in a scatter plot by minimizing the sum of the squared residuals.
• Residuals: The differences between the observed values and the values predicted by the regression model.
• Unstandardised Coefficients (B): Useful for understanding the real-world impact of predictors.
• Standardised Coefficients (Beta): Useful for comparing the relative importance of predictors.
• Simple Linear Regression: Equation: 𝑦=𝑏0+𝑏1𝑥+𝜖.
• Multiple Linear Regression: Equation not provided.