Chapter 4: Describing Distributions - Section 1: Scatterplots and Correlation

Chapter 4: Describing Distributions

Section 1: Scatterplots and Correlation

Overview: Relationships Between 2 Quantitative Variables
  • The exploration of relationships between two quantitative variables is essential in statistics.
  • We have previously examined individual categorical and quantitative variables, but understanding the relationship between two quantitative variables is crucial.
  • Key tools for analyzing these relationships include scatterplots, correlation, and regression analysis.
Scatterplot
  • A scatterplot is the most common method to display the relationship between two quantitative variables.
  • In a scatterplot:
    • Independent Variable (Explanatory Variable): Plotted on the x-axis.
    • Dependent Variable (Response Variable): Plotted on the y-axis.
  • Each point on the scatterplot represents an individual data point corresponding to the two variables.
Examples of Variables
  1. The amount of gas left in a vehicle's tank vs. the distance traveled by the vehicle.
  2. The amount of time spent studying vs. the grade received on an exam.
New Vocabulary
  • Explanatory Variable: The variable that explains changes seen in another variable; considered the independent variable (often referred to as x).
  • Response Variable: The variable that is affected by the changes in the explanatory variable; considered the dependent variable (often referred to as y).
Challenges in Determining Relationships
  • It should be noted that not all scatterplots will reveal a clear relationship between the two variables being analyzed.
Interpreting a Scatterplot
  • When interpreting scatterplots, the acronym DUFS is used to describe distributions:
    • Direction: The trend of the data points (indicates whether the relationship is positive, negative, or neutral).
    • Unusual Features: Comments on outliers, clusters, and influential points in the data.
    • Outliers: Data points that deviate significantly from the rest of the data.
    • Clusters: Groups of data points that occur more closely together than the rest.
    • Influential Points: Specific data points that can significantly affect the slope or position of the regression line.
    • Form: The general shape of the relationship (can be linear, nonlinear).
    • Strength: The degree to which the data points cluster around a line (strong, weak, or none).
Direction
  • Negative: As one variable increases, the other variable decreases.
    • Example: Time spent on a task vs. completion time could show a negative correlation in certain contexts.
  • Positive: Both variables increase together or both decrease together.
    • Example: The amount of gas consumed vs. distance traveled; both variables may increase together as consumption rises with distance.
Unusual Features
  • Unusual features in scatterplots could include the presence of outliers, which are data points significantly outside the overall pattern of the other points in the scatterplot.
  • It's important to analyze these features to draw accurate conclusions from the data.
Form
  • Linear: Indicates a straight-line relationship between the two variables.
  • Nonlinear: Indicates a relationship that does not follow a straight line; may require more complex models to describe the relationship accurately.
Strength
  • Strong: Data points are closely clustered around a straight line, indicating a strong correlation between the independent and dependent variables.
  • Weak: Data points show some direction but are more scattered, indicating a weak correlation.
  • None: No discernible direction or relationship can be detected from the scatterplot; points appear random.
Example Scatterplot Data
  • Variable Y (Dependent): 100, 60, 40, 0
  • Variable X (Independent): 0, 10, 20, 30
  • Illustrative data points that indicate a possible relationship could be included in a plot to visualize scatter and correlation.
Using a Graphing Calculator to Create a Scatterplot
  1. Navigate to the Stat menu -> Edit: Input X values into list L1 and Y values into list L2.
  2. Choose the type of graph to plot (example: y = choose #1).
  3. Navigate to Graph menu to display the scatterplot based on your data.
  4. Use Zoom menu -> select #9 for a proper scatter plot display.