Solution
We can assume a normal distribution since the problem states that the samples came from this distributoon
And for this case we can use the following formula:
[tex]\operatorname{mean}\pm Z_{\frac{\alpha}{2}}\cdot\frac{\sigma}{\sqrt[]{n}}[/tex]For this case mean = 40 , sigma = 7, n=13 and z_alpha/2 represent a quantile from the normal standard distribution, for the confidence levels 90%, 95% and 99% the z values are: 1.64, 1.96 and 2.58 and replacing we have:
90% confidence
[tex]40\pm1.64\cdot\frac{7}{\sqrt[]{13}}=(36.816;43.184)[/tex]95% confidence
[tex]40\pm1.96\cdot\frac{7}{\sqrt[]{13}}=(36.195;43.805)[/tex]99% confidence
[tex]40\pm2.58\cdot\frac{7}{\sqrt[]{13}}=(34.991;45.009)[/tex]