Chapter 18

Looking for Links: Correlation and Regression

In This Chapter

  • Exploring statistical relationships between numerical variables
  • Looking at correlation and linear regression
  • Making predictions based on known relationships
  • Considering correlation versus causation

Today's media provide a steady stream of information, including reports on all the latest links that have been found by researchers. Just today I heard that increased video game use can negatively affect a child's attention span, the amount of a certain hormone in a woman's body can predict when she will enter menopause, and the more depressed you get, the more chocolate you eat, and the more chocolate you eat, the more depressed you get (how depressing!).

Some studies are truly legitimate and help improve the quality and longevity of our lives. Other studies are not so clear. For example, one study says that exercising 20 minutes three times a week is better than exercising 60 minutes one time a week, another study says the opposite, and yet another study says there is no difference.

If you are a confused consumer when it comes to links and correlations, take heart; this chapter can help. You'll gain the skills to dissect and evaluate research claims and make your own decisions about those headlines and sound bites that you hear each day alerting you to the latest correlation. You'll discover what it truly means for two variables to be correlated, when a cause-and-effect relationship can be concluded, and when ...

Get Statistics For Dummies®, 2nd Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.