Correlations

# Correlations - Interpretation Correlations give a value...

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

Correlations There are two ways of finding a correlation, depending on the type of questions you asked. 1) Pearson’s r This is used to correlate two numerical values (like age and weight or height and weight) You can download an Excel template that does this correlation for you: http://www.dkstatisticalconsulting.com/statistics-resources/ 2) Kendall’s tau This is used when at least one of the variables is “ordinal” or uses a rating scale (such as Likert scales, or rating something 1-5). There is an online tool for this here: http://www.wessa.net/rwasp_kendall.wasp For each of these correlations, you must manually enter data. This takes some time, but yields a very useful statistical conclusion.
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Interpretation Correlations give a value between -1 and 1. Negative values only mean that as one variable increases, the other decreases (such as alcohol consumption and grades). Negative correlations are still correlations and still indicate a pattern. < .3 = no correlation ~.3-.4 = weak correlation ~.5-.6 = moderate correlation .7 or greater = strong correlation After finding and reporting your correlation, you should try to explain why you think it occurs. Why do grades decline as alcohol use rises? Why does weight increase as height increases? You should be able to offer a logical conclusion....
View Full Document

## This note was uploaded on 03/03/2011 for the course ENG 11011 taught by Professor Dr.susant.lord during the Spring '08 term at Kent State.

Ask a homework question - tutors are online