RobertNiles.com
About Robert
Reporting Help
Finding Data on the Internet
Questions for Candidates
Stats Lessons
Mean
Median
Percent
Per capita
Standard Deviation
Margin of Error
Data Analysis
Sample Sizes
Stats Tests
Help Board
Bookstore


A Mathematician Reads the Newspaper

Read about reporters who blew it when trying to write about stats. A mathematician leads you through the tricks, citing contemporary examples.
More information
-->

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

bivariate versus multivariate
Message posted by Joan (via 146.203.198.85) on August 30, 2001 at 10:12 AM (ET)

Can a variable not correlated with the dependent variable in bivariate test ever have a significant relationship with the dependent variable when you include all the other variables you think should influence the outcome and run the multivariate model?
>


READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: bivariate versus multivariate
Message posted by Jack Tomsky (via 208.249.113.130) on August 30, 2001 at 1:29 PM (ET)

The partial correlation between a pair of variables, conditional on a set of other variables, can definitely be nonzero, even if the simple correlation between the pair is zero. This can be seen by looking at the equation for partial correlations.

The partial correlation can be interpreted as the correlation between the residuals when you regress each of the two variables against the other set of variables.



Your $5 contribution helps cover part the $500 annual cost of keeping this site online.

Niles Online Sites:RobertNiles.comTheme Park InsiderViolinist.com

RobertNiles.com™, the site, content and services 咀opyright 1996-2002, Robert Niles.
All rights reserved. Questions? Comments? Read my Privacy Policy, or E-mail me!