Summarizing and Graphing Data

Frequency Distributions

  • Lower Class Limits: The smallest numbers that can belong to a class.
  • Upper Class Limits: The largest numbers that can belong to a class.
  • Class Boundaries: Values that split the difference between the end of one class and the beginning of the next.
  • Class Midpoints: The middle value of each individual class.
  • Class Width: The difference between two consecutive lower class limits.
  • Cumulative Frequency Table: Displays a running total of frequencies for a class combined with all previous classes.
  • Relative Frequency Table: Replaces raw frequencies with proportions or percentages for each class.
  • Steps to Create a Frequency Distribution:
    • Determine the number of classes (using guidelines such as Sturges' guideline).
    • Calculate class width.
    • Set lower class limits and upper class limits.
    • Tally sample values into respective classes.

Histograms and Data Distributions

  • Histogram: A visual display of a frequency distribution using bars of equal width.
    • Horizontal scale represents quantitative data classes.
    • Vertical scale represents frequencies and must start at zero.
  • Common Distribution Shapes:
    • Bell-Shaped (Normal) Distribution: Data values cluster in the middle and taper off symmetrically toward either extreme.
    • Uniform Distribution: Data frequencies are approximately equal across all classes.
    • Skewed to the Right (Positively Skewed): Distribution tail extends to the right.
    • Skewed to the Left (Negatively Skewed): Distribution tail extends to the left.

Distribution Shapes including Normal, Uniform, Skewed Right, and Skewed Left

  • Outliers: Sample values that lie very far away from the vast majority of other sample values.

Graphical Displays and Misleading Graphs

  • Types of Graphs:
    • Bar Graph: Uses bars of equal width to display frequencies or relative frequencies of categorical data.
    • Pareto Chart: A bar graph for categorical data arranged in descending order of frequency.
    • Pie Chart: Displays qualitative data as slices of a circle proportional to class frequencies, where slice angle is calculated as angle=frequencytotal×360∘\text{angle} = \frac{\text{frequency}}{\text{total}} \times 360^\circ.
    • Time-Series Graph: A line graph displaying data collected at consecutive points in time (e.g., monthly or yearly).
    • Frequency Polygon: A line graph constructed by connecting the midpoints of the tops of histogram bars; both ends extend to the horizontal axis.
    • Ogive: A line graph depicting cumulative frequencies to show how many values fall below a given limit.
    • Scatterplots, Stemplots: Additional visual displays for quantitative data distribution analysis.
  • Sources of Graphic Deception:
    • Nonzero Axis: Truncating a vertical axis by not starting at zero exaggerates differences between categories.
    • Pictographs and Disproportionate Scale: Using 3D images or scaled symbols distorts linear comparisons into area or volume misconceptions.

Comparison of vertical scale starting at 30 versus starting at zero