I'm writing an article on pyelonephritis in children and for this I've come across a statistical problem, seeing as my statistical skills are very limited.
My data is devided into three groups: group A: children younger than 6 months, Group B: Children between 6 months and 3 years, group C: children older than 3 years.
For these three groups I've found frequences of abnormal lab findings, abnormal ultrasound etc.
In order to compare these three groups statistically - as in: is the frequencies of abnormal lab findings or abnormal ultrasound significantly different? - I've been told I need to use af Chi square test with two degrees of freedom.
Can anybody tell me where I can find a calculater for this or help me in any other way.
thanks in advance for any help
Jun 19th 2009, 01:13 PM
I think doing a Chi-Test can be pretty tricky if you donīt know statistics.
There are calculators that can do this, and software, but you still need to know how to input the data etc.
This was something I found, it might help you.
There are plenty of resources online.
The Chi-Test is used to check wheter data contradicts the choice of a specific distribution function.
Other tests, like a deviance test, is often used for checking if different data are simply variance, or if some data actually have a different underlying distribution, or at least other parameters in the distribution function.
Jun 19th 2009, 09:58 PM
Thanks for your help. I'll have a look at it and see if I can figure it out.