Descriptive Statistics: Graphical Methods for Data Presentation

Chapter 2: Descriptive Statistics

Section 2.2: More Graphs and Displays

This section explores various graphical methods for displaying both quantitative and qualitative data. These visualizations help summarize, interpret, and communicate data patterns effectively.

Graphing Quantitative Data Sets

• Stem-and-Leaf Plot: A method where each data value is split into a 'stem' (all but the final digit) and a 'leaf' (the final digit). This plot is similar to a histogram but retains the original data values, making it useful for sorting and identifying patterns.

• Dot Plot: Each data value is represented by a dot above a number line. Dot plots are useful for small to moderate-sized data sets and allow for easy identification of clusters, gaps, and outliers.

Example: For the data set 21, 25, 25, 26, 27, 28, 30, 36, 36, 45, a stem-and-leaf plot and a dot plot both reveal that most values are clustered between 20 and 50.

Variations of Stem-and-Leaf Plots

• Using two rows per stem (e.g., one for leaves 0–4, another for 5–9) provides a more detailed view of the data distribution.

• This method can highlight finer patterns, such as concentrations within subranges.

Example: When organizing text message counts, using two rows per stem shows most users sent between 20 and 80 messages.

Graphing Qualitative Data Sets

• Pie Chart: A circular chart divided into sectors, each representing a category's proportion of the whole. The area of each sector is proportional to the category's frequency or relative frequency.

• Pareto Chart: A vertical bar graph where bars are ordered from tallest to shortest, representing frequencies or relative frequencies of categories. This chart highlights the most significant categories.

Example: Pie charts can show the distribution of earned degrees, while Pareto charts can display leading causes of death, emphasizing the most common causes.

Graphing Paired Data Sets

• Scatter Plot: Used for paired quantitative data, where each point represents an ordered pair (x, y). Scatter plots reveal relationships or correlations between two variables.

• Time Series Chart: Plots quantitative data collected at regular intervals over time. Points are connected by line segments to show trends or patterns over time.

Example: Fisher's Iris data set uses a scatter plot to show that as petal length increases, petal width also tends to increase. A time series chart of motor vehicle thefts over several years can reveal trends such as increases or decreases over time.

Key Definitions and Concepts

• Frequency: The number of times a data value occurs.

• Relative Frequency: The proportion of data values in a category, calculated as frequency divided by the total number of observations.

• Outlier: A data value that is significantly higher or lower than most other values in a data set.

• Cluster: A group of data values that are close together.

Formulas

• Central Angle for Pie Chart: $\text{Central Angle} = \text{Relative Frequency} \times 360^\circ$

• Relative Frequency: $\text{Relative Frequency} = \frac{\text{Frequency}}{\text{Total Number of Observations}}$

Applications and Interpretation

• Stem-and-leaf and dot plots are best for small to moderate quantitative data sets, allowing for quick identification of distribution shape, clusters, and outliers.

• Pie charts and Pareto charts are ideal for summarizing categorical (qualitative) data, making it easy to compare proportions and highlight the most significant categories.

• Scatter plots are essential for exploring relationships between two quantitative variables, while time series charts are used to analyze trends over time.

Table: Comparison of Graph Types

Graph Type	Data Type	Main Purpose
Stem-and-Leaf Plot	Quantitative	Show distribution and retain original data values
Dot Plot	Quantitative	Visualize frequency and identify clusters/outliers
Pie Chart	Qualitative	Show proportions of categories
Pareto Chart	Qualitative	Highlight most frequent categories
Scatter Plot	Paired Quantitative	Show relationships between two variables
Time Series Chart	Quantitative over Time	Show trends over time