Chi Square Test of Independence with MS Excel
(Page 1 of 4 )
One of the most important aspects of any type of research that includes surveys and scientific experiment is having an “unbiased” form of decision. Only when you arrive at an unbiased decision can your research be considered to have yielded "facts." Yet lots of managers, engineers and marketers fail to produce facts even with numerical data presented and analyzed by MS Excel. Fortunately, the Chi Square statistical tool can help.
In this article, you will learn how to use the Chi Square Test of Independence (CSTI) in MS Excel. This is one of the most important statistical tools that can guide any researcher to an unbiased or factual decision.
Facts vs. Inferences
If you are presented with a data table like the one below:
(Data source: http://davidmlane.com/hyperstat/B150300.html)
Say someone asks you whether the drug has some effect on chess players’ performance. What you will do is make some educated guess based on the results.
Example:
Calculating the %win under the condition of drug use:
% win = 12/40 = 30%
Compute the %win for those normal players (not using the drug):
% win = 13/30 = 43%
And respectively for the % Loss:
Using drug = 18/40= 45%
Without drug = 7/30= 23%
Finally for % Draw:
Using drug = 10/40= 25%
Not using drug= 10/30= 33%
Then you might say there is no positive effect because the %win is not good compared to the control group; however, what if someone asked if 43% is really a significant difference when compared to 30%, or is this just random chance? This is where analysis could get confusing due to lack of a systematic approach to problem solving.
This is where the Chi Square test of independence is very helpful.
Next: Setting up Contingency Tables and the Requirements >>
More Microsoft Access Articles
More By Codex-M
