ap stats sample question...

<p>im studying stats by going over past test…and im facing this question about null hypothesis and stuff…
my class just finished this chapter by we did not cover type 1 and type 2 error…can someone give me some help on this question??</p>

<p>when a law firm represents a groyp of people in a class action lawsuit and wins that lawsuit, the firm receives a percentage of the group’s monetary settlement. that settlement amout is based on the total number of people in the group- the larger the group and the larger the settlement, the more money the firm will receive.</p>

<p>a law firm is trying to decide whether to represent car owners in a class action lawsuit against the manufacturer of a certain make and model for a particular defect. if 5 percent or less of the cars of this make and model have the defect, the firm will not recover its expenses. therefore, the firm will handle the lawsuit only if it is convinced that more than 5 percent of cars of this make and model have the defect. the firm plans to take a random sample of 1000people who bought this car and ask them if they experienced this defect in their cars.</p>

<p>a.define the parameter of interest and state the null and alternative hypotheses that the law firm should test.
H0: P=.05
H1: P>=.05
i know this but i dont know what to do with “define the parameter of interest…”</p>

<p>b.in the context of this situation, describe type 1 and type 2 errors and describe the consequences of each of these for the law firm…
now im lost…</p>

<p>Parameter of interest is just what aspect of the population you are interested in (mean, proportion, standard deviation, etc).</p>

<p>As for type 1 and 2 errors, when doing these inference tests. There are four possible outcomes in the end. You will either

  1. Reject the null and it turns out the null was false (Correct decision)
  2. Fail to reject the null and it turns out the null was true (Correct decision)
  3. Reject the null and it turns out the null was true (Incorrect decision, this is a Type 1 error)
  4. Fail to reject the null and it turns out the null was false (Incorrect decision, this is a Type 2 errors)</p>

<p>The way to remember this is that Type 1 errors are usually the worser of the two to make (such as when testing for a disease and the test comes back negative and the person actually has it)</p>