One tailed confidence interval $1 - 2\alpha $ rationale In my statistics textbook, I’ve noticed a pattern that is clear but the rationale for which isn’t explained:
When constructing a one tailed confidence interval, the confidence level is equal to $ 1 - 2\alpha $. I can apply this idea, but can someone please explain to me the rationale? Why is $ 1 - 2\alpha $ the confidence level for a one tailed test when a two tailed test is $1 - \alpha $? 
I should add that my understanding is that a one tailed confidence interval should have confidence level equal to $1 - \alpha $ because each tail in a two tailed confidence interval has area $1 - \alpha/2 $. This is what I cannot reconcile. 
Edit: This table is referenced repeatedly in the book but no justification I can find is given. The book is “Essentials of Statistics,” 5th edition, Triola, Mario F.

Edit: I’ve included some reference photos from two sections of the book. The first two photos below are from a general section on confidence intervals where only two tailed intervals are discussed. 

The following two photos are from a section on chi-squares tests. I can not reconcile the statements regarding confidence level in the next photo from the statement about alpha in the final photo:


 A: Given the significance level $\alpha$, the confidence level must be the same $(1-\alpha)\cdot 100\%$ for both one-tailed and two-tailed hypothesis tests, but the critical values must differ. Here is an alternative table of the confidence levels:
$$\begin{array}{c|c|c|c}
{\text{Sign.level}\\ \ \ \ \ \ \alpha} & {\text{Conf.coef}\\ \ \ \ \ 1-\alpha} & { \ \ \ \ \ \text{Conf.level}\\(1-\alpha)\cdot 100\%} & {\text{Crit.value}\\ \ \ \ \ \ \ \ z_{\alpha}} & {\text{Crit.value}\\ \ \ \ \ \ z_{\alpha/2}}\\
\hline
0.01 & 0.99 & 99\% & 2.33 & 2.575 \\
0.05 & 0.95 & 95\% & 1.64 & 1.96 \\
0.10 & 0.90 & 90\% & 1.28 & 1.64
\end{array}
$$
So, in the referenced "Confidence interval method" table the one-tailed confidence level $98\%$ (corresponding to $\alpha=0.01$) must be $99\%$. It   seems it tries to convert one-tailed test to a confidence interval (two-tailed test), which is not appropriate, because the confidence interval (corresponding to one-tailed test) will be one-sided (depending on the alternative hypothesis). See this for reference. 
