This is my take-home homework. I know how to input the data to the Rguroo. But, I'm confused what the test I have to use?
Can anyone can tell me what I need to do?
By convention, a study should have a statistical power of at least 80% (0.8) to be considered a "good study." However, it is not uncommon for studies to be published despite having much lower power.
The file rheumatoid_arthritis_power.csv contains the following variables for 34 studies of risk factors for rheumatoid arthritis:
· Marker: the genetic or lifestyle risk factor being studied
· Year: the year the meta-analysis reporting the study was published
· Author: the first author of the meta-analysis reporting the study
· Power: the estimated power of the study investigating the marker (reported as a proportion)
You can consider Year and Author to jointly describe the "source" of the information used to estimate the power for each study.
Assuming that this sample is representative of all studies published about risk factors for rheumatoid arthritis, do rheumatoid arthritis studies, on average, achieve 80% power?
Important Note: This question does not ask you to compute power. Power is the variable of interest.
A) What inferential procedure would you use to answer this question?
B) Perform background and exploratory data analysis to determine whether the assumptions of the procedure have been met. Write and paste your analysis below.
C) Are all assumptions met (or at least not terribly violated)? If not, explain the violation(s).
D) If assumptions are violated, stop here. If assumptions are okay, perform the inference: