Math Is Fun Forum
  Discussion about math, puzzles, games and fun.   Useful symbols: ÷ × ½ √ ∞ ≠ ≤ ≥ ≈ ⇒ ± ∈ Δ θ ∴ ∑ ∫ • π ƒ -¹ ² ³ °

You are not logged in.

#1 2015-12-28 15:35:27

MathsIsFun
Administrator
Registered: 2005-01-21
Posts: 7,664

Chi-Square Test

Have finished Chi-Square Test and Chi-Square Calculator

Please cast your eyes over them both, and let me know any errors or improvements so visitors get the best possible experience.


"The physicists defer only to mathematicians, and the mathematicians defer only to God ..."  - Leon M. Lederman

Offline

#2 2015-12-28 17:45:50

bobbym
bumpkin
From: Bumpkinland
Registered: 2009-04-12
Posts: 109,606

Re: Chi-Square Test

Hi MIF;

Your examples work out and the page is good but I have a question:

This test only works for categorical data (data in categories), such as Gender {Men, Women} or color {Red, Yellow, Green, Blue} etc, but not numerical data such as height or weight.

I have used chi square on other than categorical data, for instance, testing the validity of a die by rolling it 6000 times and counting up the number of times 1 to 6 comes up. Say that the answer looks like this,{900,1100,1200,1000,900,900} you can calculate a p value for this to see whether or not this result could be by chance or not. So, does the quote still apply to n x 1 data? Would you consider my data as categorical?


In mathematics, you don't understand things. You just get used to them.
If it ain't broke, fix it until it is.
Always satisfy the Prime Directive of getting the right answer above all else.

Offline

#3 2015-12-28 18:02:51

ganesh
Administrator
Registered: 2005-06-28
Posts: 23,460

Re: Chi-Square Test

Hi MathsIsFun,

The pages Chi-Square Test and Chi-Square Calculator are well made! Thanks!


It is no good to try to stop knowledge from going forward. Ignorance is never better than knowledge - Enrico Fermi. 

Nothing is better than reading and gaining more and more knowledge - Stephen William Hawking.

Offline

#4 2015-12-29 12:04:48

MathsIsFun
Administrator
Registered: 2005-01-21
Posts: 7,664

Re: Chi-Square Test

Thanks guys!

bobbym - did you compare {900,1100,1200,1000,900,900} to {1000,1000,1000,1000,1000,1000} ? Otherwise Chi-square is 0 (and DF 0 also)

If so I get a Chi-square of 38.7332 and p = 2.68247e-7 , which is quite cool as it shows that to be a very expected result.

So, we can agree here to give them honorary categories of test1 to test6 and random vs boring.


"The physicists defer only to mathematicians, and the mathematicians defer only to God ..."  - Leon M. Lederman

Offline

#5 2015-12-29 14:18:47

bobbym
bumpkin
From: Bumpkinland
Registered: 2009-04-12
Posts: 109,606

Re: Chi-Square Test

Hi MIF;

Yes, I would compare them to {1000,1000,1000,1000,1000,1000}. Thanks for the explanation.


In mathematics, you don't understand things. You just get used to them.
If it ain't broke, fix it until it is.
Always satisfy the Prime Directive of getting the right answer above all else.

Offline

#6 2016-02-06 14:50:57

mathaholic
Member
From: Earth
Registered: 2012-11-29
Posts: 3,251

Re: Chi-Square Test

That was kind of nosebloodening (hard for me to understand). But... Thanks for the info! big_smile


Mathaholic | 10th most active poster | Maker of the 350,000th post | Person | rrr's classmate
smile

Offline

Board footer

Powered by FluxBB