In a book, from a sample they derived Mantel-Haenszel chi-square statistic $$\chi_1^2=1.41$$
And it is written that :
this $\chi_1^2=1.41$ is associated with a one-sided P-value between $0.10$ and $0.15$
I tried to compute the P-value . But My result is not $0.10$ and $0.15$. So i doubt whether my procedure to compute the P-value is appropriate.
$$(1/2)pr(\chi_1^2\ge 1.41|H_0)\approx pr(Z>\sqrt(1.41)|H_0)=0.1175282$$
- How can i calculate the P-value and why is it one-sided?Is that for $\chi^2$ is a one-tail test?
The probability that a chi-squared random variable with one degree of freedom obtains a value higher than $1.41$ is $.235$,
$$P\left(\chi^2_1\geq 1.41\right) = 0.235$$
So what the authors say is valid only if they are conducting a two-tailed test, and so this probability is split in two (which leads to $0.1175$). And in any case, the phrasing they use is confusing.