In a study of treatments for very painful​ "cluster" headaches, 140 patients were treated with oxygen and 158 other patients were given a placebo consisting of ordinary air. Among the 140 patients in the oxygen treatment​ group, 113 were free from headaches 15 minutes after treatment. Among the 158 patients given the​ placebo, 35 were free from headaches 15 minutes after treatment. Use a significance level to test the claim that the oxygen treatment is effective. A) Find test statistic z B) Find the P-value C) Construct the appropriate confidence interval D) determine if the oxygen treatment is effective

Respuesta :

Answer:

A

  [tex]t = 10.1[/tex]

B

  [tex]p-value = p(t > 10.1)= 0.000[/tex]

C

  [tex]0.4714 < p_1 - p_2 <0.6988[/tex]

D

  The  oxygen treatment  is effective because

1 the is no 0 in the interval telling us that the treatments are different

 2 the upper and the lower limit are positive tell us that the proportion of those treated by the oxygen treatment is greater

Step-by-step explanation:

From the question we are told that

    The  first sample size is  [tex]n_1 = 140[/tex]

     The  number of patient which the oxygen cured is k = 113  

     The second sample size is  [tex]n_2 = 158[/tex]

     The number of patient that  placebo cured is  l =  35

The first sample proportion is

           [tex]\r p_1 = \frac{ 113}{140 }[/tex]

           [tex]\r p_1 = 0.8071[/tex]

The second sample proportion is  

           [tex]\r p_2 = \frac{ 35}{ 158 }[/tex]

          [tex]\r p_2 = 0.222[/tex]

The null hypothesis is  [tex]H_o : p_1 = p_2[/tex]

 The  alternative hypothesis is  [tex]H_a : p_1 > p_2[/tex]

Let assume the level of significance be[tex]\alpha = 0.05[/tex]

Generally the pooled proportion is mathematically evaluated as

    [tex]p = \frac{p1 * n1 + p2 * n2}{n1 + n2}[/tex]

substituting values

     [tex]p = \frac{0.8071 * 140 + 0.222 * 158}{140 + 158}[/tex]

      [tex]p = 0.4969[/tex]

Generally the standard error is mathematically represented

      [tex]SE = \sqrt{ p(1- p ) * [ \frac{1}{n_1} + \frac{1}{n_1}] }[/tex]

      substituting values

     [tex]SE = \sqrt{ 0.4969(1- 0.4969 ) * [ \frac{1}{140} + \frac{1}{158}] }[/tex]

     [tex]SE = 0.0580[/tex]

Generally the test statistics is mathematically represented as

       [tex]t = \frac{ \r p_1 - \r p_2}{ SE}[/tex]

       [tex]t = \frac{ 0.8071 -0.222}{ 0.0580}[/tex]

       [tex]t = 10.1[/tex]

The  p-value is from the normal distribution table as  

      [tex]p-value = p(t > 10.1)= 0.000[/tex]

given that [tex]t< \alpha[/tex] the null hypothesis is rejected

From the normal distribution table the critical value of  [tex]\frac{\alpha }{2}[/tex]  is  

     [tex]Z_{\frac{\alpha }{2} } = 1.96[/tex]

Generally the margin of error is mathematically the represented as

         [tex]E = Z_{\frac{\alpha }{2} } * SE[/tex]      

         [tex]E = 1.96 * 0.0580[/tex]      

          [tex]E = 0.1137[/tex]

The 95% confidence interval is mathematically  represented as

           [tex](\r p_1 - \r p_2) - E < p_1 - p_2 < (\r p_1 - \r p_2) + E[/tex]

substituting value

             [tex](0.8071 - 0.222) - 0.1137 < p_1 - p_2 < (0.8071 - 0.222) + 0.1137[/tex]

             [tex]0.4714 < p_1 - p_2 <0.6988[/tex]

The  oxygen treatment  is effective because

1 the is no 0 in the interval telling us that the treatments are different

 2 the upper and the lower limit are positive tell us that the proportion of those treated by the oxygen treatment is greater