Chapter 13

  1. The reasoning for saying “provided strong evidence that using the busy road was quickest” as opposed to “proved that using the busy road was quickest” is because the outcome of the experiment cannot be “proved” without measureable data.

  2. A very small p-value was found despite the difference in travel times between the two raoutes being trivial because p-values are for hypothetical testing and are not error rates. P-values allow researchers to narrow possibilities.

  3. If statistics is not soleley for testing hypotheses, this field can be used for the collection, analysis, interpretation and presentation of data.

Chapter 14

  1. We say probability of the null hypothesis rather than that of the hypothesis given the data because we are able to only reject or not reject the null hypothesis based off the p-value data. The p-value essentially is the probability of receiving results given that the null hypothesis is true.

  2. An example of the null hypothesis for each of these questions would be: “Do African American males have a harder time than white males hailing a taxi in New York City?” “Nationwide, about 28% of births are via Cesarean delivery. Do hosipitals in New York state have higher than average Cesarean rates?” “Do patitients taking a new, less toxic type of chemotherapy have response rates at least as agood as those on the standard chemotherapy drug?

b) African American males have an equal chance of hailing taxis in New York City c) Hospitals in New York state have average Cesarean section rates. e) Patients taking new, less types of chemo have response rate equal to those on standard chemo drugs.

Chapter 15

  1. It is not always true that a p-value of 0.049999 is always statistically signficant and that one with a value of 0.050001 is never significant. The size of sample size will determine whether or not they are significant.

  2. I think the experiment with Michael Jordan resulted in an insignificant p-value because the sample size that was used in the experiment was not large enough.

  3. The connection between a criminal trial and a p-value is the two types of errors that may come out of a criminal trial Type I error in this instance would be sending an innocent person to jail and Type II would be letting a guilty person be set free.