Discussion (WK5)

Discussion (WK5) - In case of correlation we do not have to...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Discussion (WK5) Question 1: What is correlation? Does correlation prove causation? Why or why not? Explain and provide examples to support your explanation. Correlation signifies the extent to which there exists a proportional change relationship between two variables. Correlation does not necessarily prove causation because correlation may at times be spurious or co-incidental and not always based on cause and effect. An example of spurious correlation may be the existence of positive correlation between the marks of an English test of students and their weights. Reference: th Ed. McGraw-Hill Irwin. 2011: CH13, Pg 385. Questions 2: What are the differences between regression and correlation analysis?
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: In case of correlation, we do not have to think about cause and effect relationship whereas for regression, we do have to think about cause and effect relationship as it predicts Y from X. In correlation, any variable can be Y and X whereas in regression the line predicts Y from X is not same as X from Y. Correlation is rarely appropriate when one variable is something which you experimentally manipulate. And for regression X variable is usually something you experimentally manipulate and Y variable is something you measure. Reference: Lind, Marchal, Wathen. Basic Statistics for Business & Economics, 7 th Ed. McGraw-Hill Irwin. 2011: CH13, Pg 385....
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online