Bivariate Data Analysis: Categorical and Quantitative Relationships

Bivariate Data

Data involving two variables used to determine associations between them.

Two-Way Table

A table used to organize counts for two categorical variables.

Marginal Distribution

The distribution of one variable disregarding the other, often found in the totals of a two-way table.

Conditional Distribution

The distribution of one variable restricted to a specific category of another variable.

Independence of Variables

Two variables are independent if the conditional distribution of one variable remains the same across all categories of the other variable.

Association of Variables

Occurs when knowing the value of one variable helps predict the value of the other.

Side-by-Side Bar Chart

A graphical representation where bars for different groups are placed next to each other for comparison.

Segmented Bar Chart

A bar chart divided into segments that represent percentages of a second variable.

Mosaic Plot

A variation of a segmented bar chart where the width of the bars is proportional to the sample size.

Scatterplot

A graphical representation that plots the explanatory variable against the response variable.

Direction in Scatterplots

Refers to whether the relationship is positive (uphill) or negative (downhill).

Unusual Features in Scatterplots

Includes outliers or distinct clusters that deviate from the overall pattern.

Form of Relationship

Describes the shape of the relationship depicted in the scatterplot (e.g., linear, nonlinear).

Strength in Scatterplots

Indicates how closely the points fit a specific form, such as being weak, moderate, or strong.

Correlation Coefficient ($r$)

A measure of the strength and direction of a linear relationship between two quantitative variables.

Properties of $r$

Range is between -1 and 1, where values near ±1 indicate strong relationships, values near 0 indicate weak relationships.

Least Squares Regression Line (LSRL)

The line that minimizes the sum of the squared residuals in a regression analysis.

Residual

The difference between an observed value and the predicted value in a regression model.

Positive Residual

Indicates that the observed value is above the predicted value.

Negative Residual

Indicates that the observed value is below the predicted value.

Residual Plot

A graph that displays the residuals on the y-axis and the explanatory variable on the x-axis.

Coefficient of Determination ($r^2$)

Represents the proportion of variation in the dependent variable that is explained by the model.

Outlier

A data point with a large residual that weakens the correlation.

High Leverage Point

A point with an x-value far from the mean of x, which can influence the slope of the regression line.

Influential Point

A data point that significantly changes the slope, intercept, or correlation if removed.

Transformations in Regression

Methods used to achieve linearity if a residual plot indicates a nonlinear relationship.

Exponential Model Transformation

Involves plotting x against ln(y) to identify a potential exponential relationship.

Power Model Transformation

Involves plotting ln(x) against ln(y) to identify a potential power relationship.

Sum of Residuals

Always equals zero for the least squares regression line.

Standard Deviation of Residuals ($s$)

Measures the average distance of actual data points from the regression line.

Context in Describing Relationships

Always include descriptions related to the specific variables being analyzed.

Extrapolation

The act of predicting values outside the observed range of data, which can lead to inaccuracies.

Bar Charts vs. Histograms

Bar charts are for categorical data and show gaps between bars, while histograms are for quantitative data and do not have gaps.

Slope ($b$) Interpretation

Describes how the predicted response variable changes with a one-unit increase in the explanatory variable.

Y-Intercept ($a$) Interpretation

Describes the predicted value of the response variable when the explanatory variable is zero, if applicable.

Linear Relationship Requirements

Two quantitative variables must show a consistent pattern of change to be considered a linear relationship.

Causation

Denoted that correlation does not imply that one variable causes changes in the other.

Prediction

Using regression analysis to estimate the value of the response variable based on values of the explanatory variable.

Computational Output in Regression

Software output that provides coefficients, standard deviation of residuals, and the coefficient of determination.

Random Scatter in Residual Plots

Indicates that a linear model is appropriate for the data.

Curved Pattern in Residual Plots

Indicates that the relationship in the original data is nonlinear.

Fanning in Residual Plots

Indicates that prediction errors increase as the explanatory variable increases, suggesting model inadequacy.

Residual Analysis

Essential step to verify the appropriateness of a linear regression model by examining the distribution of residuals.

Total for Two-Way Table

The overall count of responses, found by summing up the counts of all categories in the table.

Participants in Cuteness Study

250 volunteers who measured their focus levels related to viewing different pictures.