Study Notes on Scatter Plots and Data Associations
Introduction to Scatter Plots
- Purpose: To explore the relationship between two variables using a visual representation.
- Definition: A scatter plot is a type of graph that represents paired or dependent data.
Key Features of Scatter Plots
- Datasets: Requires two separate datasets, plotted on a single coordinate plane.
- Axes:
- X-axis: Represents one variable.
- Y-axis: Represents the second variable.
- Coordinate Pairs: The paired data is converted into x and y coordinates to be plotted as points on the scatter plot.
Understanding Association in Scatter Plots
- Association: Refers to the relationship between the two variables.
- Possibilities:
- There may be an association.
- There may be no association.
- Analysis Factors: Trends, shape, and strength of the association are key to understanding the data represented.
Descriptors Used in Scatter Plots
- Trend: The overall direction of the data points on the scatter plot.
- Example: Reading from left to right, if points rise, the trend is increasing.
- Shape: The geometric representation of the data points.
- Example: A linear shape indicates a direct relationship (straight line).
- Strength: How closely the points fit the identified shape.
- Example: Strong association if points closely align with a trend line.
Examples of Scatter Plots
Example 1: Increasing Trend
- Trend: Increasing
- Shape: Approximately linear
- Strength: Strong (points closely follow a straight line pattern)
Example 2: Decreasing Trend
- Trend: Decreasing
- Shape: Linear
- Strength: Strong (points follow a straight line well)
Example 3: Moderate Strength Association
- Trend: Increasing
- Shape: Linear
- Strength: Moderate (points display more scatter, not tightly following the line)
Example 4: Decreasing Trend with Moderate Strength
- Trend: Decreasing
- Shape: Approximately linear
- Strength: Moderate (lines less tightly followed)
Example 5: No Association
- Observation: Data points scatter randomly without following a specific trend or shape.
Example 6: Changing Shape
- Shape: Parabolic or curved (not linear)
- Strength: Moderate (Some points deviate from the curve)
- Trend: Changing (not strictly increasing or decreasing)
Focus on Linear Relationships
- Primary Interest: Identifying variables that follow a linear shape (increasing or decreasing trends).
- Statistical Processes: Other types of relationship shapes can be analyzed but will not be the focus.
Constructing a Scatter Plot Example
- Using StatCrunch:
- Select
Graph then Scatter Plot. - Indicate the first and second variables to be analyzed.
- Click compute to generate the scatter plot.
Description of Scatter Plot Output
- Example Output:
- Association: Fairly strong positive linear association noted.
- Trend: Increasing trend observed as you read left to right.
- Real-World Context: Countries with more roller coasters contribute more funds to tsunami aid.
- Correlation Interpretation: Positive correlation indicates that as one variable increases, so does the other (roller coasters and tsunami aid).