STATS 10 Lecture Notes - Lecture 4: Streptomycin, Random Assignment
Course CodeSTATS 10
This preview shows half of the first page. to view the full 3 pages of the document.
UID: 904 652 912
Stas 10 Lec 4
TA: Min, S.
Lab #4: TB or Not TB?
Question 1: Summarize the results of Hill’s study. a.) Based on Hill’s study, does it seem that
treatment and outcome are independent or dependent (i.e. does it seem like Streptomycin is
ineffective or effective in treating tuberculosis)? Justify your answer.
a. we can see that 51 out of 55 in the treatment group received the streptomycin recovered.
38 out of 52 people recovered, of the people in the control group who received only bed
rest. It is clear that the people in the treatment group are more likely to recover, from
summary table. It seems like the people are dependent and therefore streptomycin is an
Question 2: a.) Without doing any simulations in Fathom, if treatment and outcome were
independent, what proportion of patients according to Hill’s study, would we expect to die if we
replicated the study again? b.) how many patients in the Streptomycin group would we expect to
die if treatment and outcome were independent?
a. The proportion of people who will die if the study was repeated again is about 0.168, if
treatment and outcome were independent. If both the treatment and outcome were
independent we would be able to expect about 18 people to die.
b. We would expect about 9 people to die in the streptomycin group. 0.168 x 55 (people in
streptomycin group) equals the total people who would die if treatment and outcome
You're Reading a Preview
Unlock to view full version