Confidence interval for mean, when population standard deviation is unknown:
[tex]\overline{x}\pm t_{\alpha/2}\dfrac{s}{\sqrt{n}}[/tex]
, where [tex]\overline{x}[/tex] = sample mean
n= sample size
s= sample standard deviation
[tex]t_{\alpha/2}[/tex] = Critical t-value for n-1 degrees of freedom
We assume the population has a normal distribution.
Given, n= 19 , s= 3.8 , [tex]\overline{x}=22.4[/tex]
[tex]\alpha=1-0.99=0.01[/tex]
A) Critical t value for [tex]\alpha/2=0.005[/tex] and degree of 18 freedom
[tex]t_{\alpha/2}[/tex] = 2.8784
B) Required confidence interval:
[tex]22.4\pm ( 2.8744)\dfrac{3.8}{\sqrt{19}}\\\\=22.4\pm2.5058\\\\=(22.4-2.5058,\ 22.4+2.5058)=(19.8942,\ 24.9058)\approx(19.9,\ 24.9)[/tex]
Lower bound = 19.9 years
Uppen bound = 24.9 years
C) Interpretation: We are 99% confident that the true population mean of lies in (19.9, 24.9) .