Question

At this point, we have a variety of options when choosing a test or a confidence interval. I'd like for you to walk us through your process of choosing based on what you see in the problem. What do you look for? What helps you decide what to choose?

Answer 1

For a hypothesis test we have the following procedure

1. State null and alternative hypothesis
2. Find test statistic = Eg: If testing for population mean= where the null value is what we equate the population parameter to (previously assumed value) SE is the standard error
3. Determine the level of significance.
4. Calculate the critical value the level of significance. This is like find x at a given probability Eg: In a z -test for 1 tailed C.V. = and for 2 tailed C.V. =
5. Compare the test stat and Critical value. If |T.S.| > C.V. then we reject the null hypo at the given

The statistics value will have to be obtained from the data provided or might even be provided directly. The critical values can be found using the distribution probability tables. As for the method it will be mentioned in the question that 'Test using hypothesis testing.'

For a confidence interval test we have the following procedure

1. State null and alternative hypothesis (same as the hypothesis test)
2. Determine the level of significance.
3. Find test statistic . There are three different range for different alternative hypothesis  (p is the population parameter) Then range =  Then range =  Then range =     
4. Calculation of margin of error = C.V. value * Std error (critical value will vary with the tails and can be calculate like in hypothesis testing)
5. Decision criteria: If the null hypothesis parameter value lies in the confidence interval then we would not reject the null hypothesis.

The statistics value will have to be obtained from the data provided or might even be provided directly. The critical values can be found using the distribution probability tables. As for the method it will be mentioned in the question that 'Test using confidence interval.'