Statistics for CSAI

by Albert Major

Understanding graphs and data visualization

Graphs help us visualize data effectively, making it easier to identify patterns and differences. Different types of graphs serve different purposes, and understanding when to use each type is important for accurate data interpretation.

Histograms: Visualizing continuous data

• A histogram is used to show the distribution of continuous data.
• Unlike a bar graph, the bars in a histogram touch because they represent ranges of values rather than separate categories.
• The x-axis represents the variable (e.g., height, weight, test scores), while the y-axis represents frequency (how often values in a range appear in the dataset).
• Histograms help identify key characteristics of data, including:
  • Shape of the distribution (e.g., normal, skewed, uniform, bimodal).
  • Skewness (whether data is asymmetrical toward higher or lower values).
  • Kurtosis (whether data has a broad or narrow peak and many or few extreme values).
• Example: If you collect height data from students and want to see the overall distribution, a histogram would be the appropriate visualization.

Bar Graphs: Comparing categorical data

• A bar graph is used for categorical data, meaning the data falls into distinct groups.

• Each bar represents a different category, and its height indicates the frequency of that category.

• The gaps between bars in a bar graph emphasize that the categories are distinct, unlike in a histogram where the bars are connected.

• Example: If you survey students about their favorite movie genres (Drama, Comedy, Horror, Action), you would use a bar graph to compare the number of students in each category.

• Bar graphs can also be used for ordinal data, where categories have a logical order (e.g., small, medium, large).

  

  Examples

Box Plots: Displaying data spread and outliers

• A box plot (also called a box-and-whisker plot) visualizes the spread and variability of data.
• A box plot is divided into quartiles:
  • First quartile (Q1): The value below which 25% of data falls.
  • Median (Q2): The middle value of the dataset.
  • Third quartile (Q3): The value below which 75% of data falls.
  • Interquartile Range (IQR): The difference between Q3 and Q1, representing the middle 50% of the data.
  • Whiskers: Lines extending from the box to indicate the range of most data points.
  • Outliers: Individual points outside the whiskers representing extreme values.