HW by linfengfengfz

VIEWS: 2 PAGES: 6

									Advanced Math Topics

10.5 Tests Concerning Means for
         Small Samples
  If you understood yesterday’s lesson about accepting/rejecting a company’s
  claim from a large sample, then today’s lesson should come quickly, as it is
  very similar.




If you did not understand yesterday’s lesson, today’s lesson may help because the
only difference is which chart to use in the back of the book.

 Refresher: A large sample is…
                            greater than 30.

             A small sample is…
                             less than or equal to 30.
A manufacturer claims that each bag of mixed nuts sold contains an average of
10 cashew nuts. A sample of 15 cans has an average of 8 cashew nuts with
a standard deviation of 3. Does this indicate that we should reject the
manufacturer’s claim? Use a 5% level of significance.
     Null Hypothesis, H0: μ = 10       Alternative Hypothesis, H1: μ ≠ 10
  It does not state that we are looking for strictly MORE THAN or LESS THAN 10 cashews,
  and we are only looking to reject the claim that there is an average of 10 cashews,
  therefore this is a two-tail test and we use ≠.

       (x  ) y
    t
            s                                          0.025                       0.025
                                                            t = -2.145 μ = 10   z = 2.145
       (8  10 ) 15                           Since the sample is small, we use the t-table in the
    t                                        back. The degrees of freedom is 15 - 1 = 14. You can
             3                                either draw a picture and label each rejection region
                                              or divide the significance level by 2 (because
     t = -2.58                                it is a two-tail test) to find the 0.05/2 = t0.025.



    Since the t-score is in the rejection region, our decision is that must reject
    the claim that each bag has an average of 10 cashews.
A new health pill is being used at a hospital to help fight cholesterol.
The manufacturer claims that anyone with high cholesterol who takes this pill
will lose 15 mg of cholesterol within a month. A doctor believes this is inflated,
& gives this pill to six people with high cholesterol and finds that they lose an
average of 12 mg with a standard deviation of 4 mg. Should we reject the
manufacturer’s claim? Use a 5% level of significance.
      Null Hypothesis, H0: μ = 15          Alternative Hypothesis, H1: μ < 15
  Since the doctor feels the manufacturer’s average is inflated, if we reject the claim, we will state
  that the average is less than 15 mg. Therefore this is a one-tail test.

         (x  ) y
    t
              s
                                                             0.05
                                                                t = -2.015 μ = 15
     (12  15 ) 6
  t                                              Since the sample is small, we use the t-table in the
           4                                      back. The degrees of freedom is 6 - 1 = 5. You can
                                                  either draw a picture and label the rejection region
                                                  or divide the significance level by 1 (because
      t = -1.84                                   it is a one-tail test) to find the 0.05/1 = t0.050.



    Since the t-score is in the acceptance region, we cannot reject the manufacturer’s
    claim that the average amount of cholesterol lost is 15 mg.
                                   From the HW P. 497
1) Welfare officials in a certain city claim that the average number of cases of child
abuse handled daily is 7.62. A newspaper tests this claim by selecting 10 days and they
finds the average for those days is 5.32 claims with a standard deviation 2.06. Using a
5% level of significance, should we reject the welfare officials’ claim?




     t = -3.53 Which is outside the acceptance region of t = -2.262 and t = 2.262.
     We must reject the officials’ claim.
                                      HW

P. 497 #1-8 You should get 4 two-tail tests and 4 one-tail tests.

								
To top