As part of an environmental impact study, a power company wants to estimate the difference in mean water temperature between the discharge of its plant and the offshore waters. How many sample measurements must be taken at each site to estimate the true difference between means to within .2°C with 95% confidence? Assume that the standard deviation in readings is about 1.5°C at each site and the same number of readings will be taken at each site.

Respuesta :

Using the z-distribution, it is found that 217 sample measurements should be taken at each site.

The margin of error of a z-confidence interval is given by:

[tex]M = z\frac{\sigma}{\sqrt{n}}[/tex]

In which:

  • z is the critical value.
  • [tex]\sigma[/tex] is the population standard deviation.
  • n is the sample size.

The first step is finding the critical value, which is z with a p-value of [tex]\frac{1 + \alpha}{2}[/tex], in which [tex]\alpha[/tex] is the confidence level.

In this problem, [tex]\alpha = 0.95[/tex], thus, z with a p-value of [tex]\frac{1 + 0.95}{2} = 0.975[/tex], which means that it is z = 1.96.

Estimate of the standard deviation of 1.5, thus, [tex]\sigma = 1.5[/tex].

We want the sample for a margin of error of 0.2ºC, thus, we have to solve for n when [tex]M = 0.2[/tex].

[tex]M = z\frac{\sigma}{\sqrt{n}}[/tex]

[tex]0.2 = 1.96\frac{1.5}{\sqrt{n}}[/tex]

[tex]0.2\sqrt{n} = 1.96(1.5)[/tex]

[tex]\sqrt{n} = \frac{1.96(1.5)}{0.2}[/tex]

[tex](\sqrt{n})^2 = (\frac{1.96(1.5)}{0.2})^2[/tex]

[tex]n = 216.1[/tex]

Rounding up:

217 sample measurements should be taken at each site.

A similar problem is given at https://brainly.com/question/17039768