Stats Data and Models

Richard D. De Veaux, Paul F. Velleman, David E. Bock

ISBN #9780321986498

4th Edition

1,406 Questions

16,980 Students Helped

Homework Questions

Summary

Learning Objectives

Key Concepts

Example Problems

Explanations

Common Mistakes

Summary

This chapter emphasizes the importance of comparing distributions to understand differences across groups, whether by using histograms for overall shape, boxplots for centers and spreads, or stem-and-leaf displays for detailed data. It also covers timeplots as tools for visualizing trends over time, the significance of outliers, and the benefit of re-expressing skewed data to improve interpretability. Understanding these methods is critical for making informed decisions based on data analysis.

Learning Objectives

Key Concepts

CONCEPT

DEFINITION

Analysis of Variance (ANOVA)

A collection of statistical techniques used to compare the means of three or more groups by partitioning total variability into components due to differences among groups (between-group) and within groups.

Example Problems

Example 1

The Research and Innovative Technology Administration of the Bureau of Transportation Statistics (www.TranStats.bts.gov /Data_Elements.aspx?Data=2) reports load factors (passenger-miles as a percentage of available seat-miles) for commercial airlines for every month from 2000 through September 2013. Here are histograms comparing the domestic and international load factors for this time period:

Example 2

Load factors, 2013 by season We can also compare domestic load factors for September through March versus those for April through August: Compare and contrast what these histograms show.

Example 3

Here is a display of the international load factors by month for the period from 2000 to $2013 .$ Discuss the patterns you see in this display.

Example 4

Here is a display of the domestic load factors by year. Discuss the patterns you see in this display.

Example 5

Extraordinary months Look at the boxplots by month in Exercise $3 .$ The three outliers in September, October, and November are all from the year 2001 and reflect the terrorist attacks of $9 / 11$. Do you think the data for these months should be set aside in any overall analysis of load factor patterns by month? Explain.