World Articles in Ear Nose and Throat 
Chi-Square Statistical Analysis
 

Recommended Link

 
WAENT Home Page

Information for Authors

Current & Back Issues

Submit Article or Video
 
ENT USA Website

Newsletter Subscription
 
Contact Us

Search WAENT & ENTUSA

 
A knowledge of basic statistics is important for the researcher.

Compare your data:  

Many research studies are controlled studies which attempt to compare an experimental group to a control group.   For example: An experimental group would be the group of patients that receive a new treatment as the control group would receive the standard treatment. 

For comparisons to be valid the two groups should be either selected randomly or in some studies they are matched.  The groups should be as similar as possible. 

Statistical Differences are described as the probability that the observed differences could occur by chance alone.   For a Statistically Significant Difference to occur, the probability that the observed difference could occur by chance alone should be less than one in twenty or a p value of less than 0.05. 

If the results can be grouped into a few categories, such as yes vs, no; or another example: red vs. white vs. green, then a Chi Square Test may be performed to determine if the two groups are different.   More About The Chi Square Test 

If the results are in a continuous data category such as weight or survival after a diagnosis, then a t-test may be able to determine if the two groups are statistically different by comparing the data's mean and standard deviation.   More About Paired T-Test

Advertisements

Copyright World Articles in Ear Nose and Throat, 2009     All rights reserved.
   

Advertisements