Part A (Study Design) What do we need to be true of a sample in order to generalize results to a population?
Part B (Study Design) Why does randomization of treatments to experimental units let us make causal conclusions?
Part C (Study Design) What is a sampling frame? What goes wrong if the sampling frame is missing subsets of the population?
Part D: Why does an odds-ratio of 1 indicate no association between variables?
Part E: In words, what does it mean to say 2 events are independent?
Part F: What two different ways can we use to check if 2 events are independent?
IMS - Section 2.5, Questions 3, 7, 10, 12, 15
A link to the text from which the questions come from is [here].
Open Intro Chapter 3, Questions 3.2, 3.5, 3.6, 3.8, 3.9, 3.16
Researchers recruited 451 patients with a high level of risk for strokes (when bloodflow gets cut off to the brain due to blocked blood vessels). They split these patients into two groups, a treated group that received stents (small mesh tube placed inside of vulnerable arteries) and medical management (medications, lifestyle coaching, etc.), and a control group that only received medical management. Of the 224 patients in the treatment group, 45 suffered a stroke within the first year of the study, while only 28 patients in the control group had a stroke during this time. (Hint: making a table/diagram of this data may be helpful)
Part A: Find the risk of having a stroke for someone in the treatment group and also the risk of having a stroke for someone in the control group.
Part B: Find the relative risk of having a stroke in the treatment group compared to the control group.
Part C: Interpret this relative risk value.
Part D: Find the probability of someone in our data having a stroke and being in the treatment group. Are these two events disjoint?
Part E: Find the odds of having a stroke for each group.
Part F: Find the odds ratio comparing the odds of stroke in the treatment (stent) group with the odds of a stroke in the control group.
Part G: Is there an association between stent use and the prevalence of strokes? Justify your answer using results from either Part B or Part F.
Part H: Explain your findings to Part G to someone who hasn’t taken a statistics class and has a high risk of having a stroke.