The total number of points in the exam is 100. The format is short answer. Partial credit will be given so please show your work. The test is open notes and open book. You may use your computer to look at your notes or do calculations, or you may use your calculator. I may ask questions about anything that was in the homework assignments, lectures, or book chapters, through the lecture on violation of assumptions . I especially recommend you study the following.
1) Why do samples need to be random? Be able to recognize a random or non-random sample.
2) What are precision and accuracy?
3) How do the mean, median and mode compare for distributions that are symmetric, right skewed or left skewed?
4) How do you estimate medians, percentiles, means and standard deviations?
5) How do you calculate the standard error of the mean? What is the relationship between standard errors and the sample standard deviation?
6) How do you estimate a proportion?
7) How do you calculate the standard error of a proportion?
8) How do you calculate a 95% confidence interval of the mean using the t distribution?
9) How do you interpret a confidence interval?
10) How do you calculate the following probabilities?
a) the probability of two independent events both happening
b) the probability of two mutually exclusive events both happening
11) If you have P(A|B) how do you calculate P(B|A)? (Bayes theorem)
12) How do you use the binomial test to test the hypothesis that a proportion is significantly different from 0.5?
13) How does the χ2 goodness-of-fit test work?
14) How does χ2 contingency analysis work? What are the Yates correction and the Fisher exact test?
15) When do you use the binomial distribution, and how do you interpret it?
16) When do you use the Poisson distribution, and how do you interpret it?
17) For all hypothesis tests, understand the following:
a) what is the null hypothesis?
b) what is the statistical distribution to be used to calculate the probabilities?
c) what is the test statistic?
d) how do you find the critical value and what does it mean?
e) how do you find the P value?
f) how do you interpret the P value?
18) Understand Type I and Type II error and the effect of sample size on Type I and Type II error.
19) How do you use the Z distribution to calculate normal probabilities?
20) What is the difference between a paired and 2-sample t test, and when would you use each test?
21) Given values of the means and standard deviations, be able to conduct both the paired and the 2-sample t test.
22) What is a normal Quantile plot (qqnormal plot) and how do you interpret it?
23) How are data transformations used? When would you use each transformation (log, arcsine-squareroot etc.)?
24) What are the Welch's test and the Mann-Whitney-Wilcoxon test, and how do they compare to the 2-sample t-test? Which would you use when?