Let $X$ be the number of seniors in the sample. We are presumably sampling without replacement, but $1000000$ is very large compared to the sample size, so we can assume that the distribution of $X$ is binomial, $n=1200$, $p=0.14$. So $X$ has mean $(1200)(0.14)$ and standard deviation $\sqrt{1200(0.14)(0.86)}$.
Let $Y=\frac{X}{1200}$, the sample proportion of seniors.
Then $Y$ has mean $0.14$, and standard deviation $\frac{\sqrt{(0.14)(0.86)}}{\sqrt{1200}}\approx 0.0100$.
Note that the distribution of $Y$ is very close to normal. So with probability $0.95$, $Y$ is within $(1.96)(0.0100)$ of the mean $0.14$. The nearest in the given list is $12\%$ to $16\%$.