Exploring Data with Tables and Graphs: Scatterplots, Correlation, and Regression

Chapter 2: Exploring Data with Tables and Graphs

Overview

This chapter introduces essential graphical and tabular methods for organizing, summarizing, and analyzing quantitative data in statistics. The focus is on frequency distributions, histograms, and the use of scatterplots to explore relationships between paired variables, including correlation and regression analysis.

Scatterplots, Correlation, and Regression

Scatterplots and Correlation

Scatterplots are a fundamental tool for visualizing the relationship between two quantitative variables. Correlation analysis helps determine whether and how strongly pairs of variables are related.

• Correlation: A correlation exists between two variables when the values of one variable are somehow associated with the values of the other variable.

• Linear Correlation: Linear correlation exists when the plotted points of paired data result in a pattern that can be approximated by a straight line.

Definition: Scatterplot (Scatter Diagram)

• A scatterplot (or scatter diagram) is a plot of paired (x, y) quantitative data with a horizontal x-axis and a vertical y-axis.

• The horizontal axis is used for the first variable (x), and the vertical axis is used for the second variable (y).

Example: Waist and Arm Circumference Correlation

• Observation: The distinct pattern of the plotted points suggests a correlation between waist circumferences and arm circumferences.

• Application: Such a scatterplot can help identify whether larger waist sizes tend to be associated with larger arm circumferences.

Example: Weight and Pulse Rate Correlation

• Observation: The plotted points do not show a distinct pattern, indicating no apparent correlation between weights and pulse rates.

• Application: This suggests that knowing a person's weight does not help predict their pulse rate.

Linear Correlation Coefficient (r)

The linear correlation coefficient, denoted by r, measures the strength and direction of the linear association between two variables.

• Range: The value of r is always between -1 and 1.

• Interpretation:

  • If r is close to -1 or 1, there appears to be a strong linear correlation.

  • If r is close to 0, there does not appear to be a linear correlation.

Formula:

$r = \frac{\sum (x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum (x_i - \bar{x})^2 \sum (y_i - \bar{y})^2}}$

Example: Shoe Print Lengths and Heights

• Paired data such as shoe print length and height can be analyzed for correlation using r.

• For a sample size of 5, a computed r value of approximately 0.59 suggests a moderate positive correlation.

• For a larger sample (n = 40), a computed r value of 0.813 with a P-value of 0.000 indicates a strong, statistically significant linear correlation.

P-Value in Correlation Analysis

The P-value is the probability of obtaining paired sample data with a linear correlation coefficient at least as extreme as the one observed, assuming no actual correlation exists.

• Decision Rule: There is sufficient evidence of significance if the P-value is equal to or less than 0.05 (5%).

• Application: A small P-value (e.g., 0.000) provides strong evidence for a significant linear correlation.

Regression Analysis

Regression is a statistical method for modeling the relationship between a dependent variable and one or more independent variables. The regression line (or line of best fit) is the straight line that best fits the scatterplot of the data.

• Regression Equation: The equation of the regression line is given by:

$y = b_0 + b_1 x$

• Where y is the dependent variable, x is the independent variable, b_0 is the y-intercept, and b_1 is the slope.

• Application: For example, predicting height from shoe print length using the regression equation.

Example: Regression Line for Shoe Print Length and Height

• Regression equation: $\text{Height} = 125.40733 + 1.7274524 \times \text{Shoe Print Length}$

• This equation allows estimation of a person's height based on their shoe print length.

Summary Table: Correlation and Regression Concepts

Concept	Definition	Key Formula	Interpretation
Correlation	Association between two variables	$r$	Strength and direction of linear relationship
Scatterplot	Graph of paired (x, y) data	—	Visualizes relationship
Regression	Modeling relationship between variables	$y = b_0 + b_1 x$	Predicts value of dependent variable
P-value	Probability of observed correlation under null hypothesis	—	Assesses statistical significance

Additional info: The notes above expand on brief slide points to provide full definitions, formulas, and context for each concept. Examples and applications are included to illustrate statistical reasoning and interpretation.