0
$\begingroup$

Is it possible to do this with chi square?

I have two groups of mice, one with a mutation in a gene and one without. I have analysed the presence/absence of organs in each group, and found that all of the non-mutated mice have all organs present, whilst the mutants are all missing the kidney (and a lung, but likely an anomaly as this was only seen in one mouse).

So the data would display the presence of the organ as yes/no. I've been asked to suggest a statistical test to analyse this data, and suggest challenges you may face (we do not actually have to carry out the analysis).

From what I've read, the chi square is recommended but I don't really understand why? Or what challenges would be faced in analysing this data in this way?

Any help is greatly appreciated, as statistics are still quite new to me

$\endgroup$
  • $\begingroup$ Is this a class assignment? $\endgroup$ – Sal Mangiafico Oct 25 at 11:32
0
$\begingroup$

One would need to know how many mice are in each group in order to give a numerical answer to your question.

If 4 mice in each group, then under $H_0,$ the probability all defective mice are in the mutant group is ${4\choose 4}{4\choose 0}/{8 \choose 4} = 1/70<0.05.$

This is an example of Fisher's exact test for the 'Lady tasting tea.'

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.