Sources of Variation Flashcards
0
Q
How do you test a hypothesis?
A
Calculate probability of getting a result as extreme as, or more extreme than, the one observed - assuming the hypothesis is true. t
This is the p value; if it is very low (normally p<0.05) the results are significant and the hypothesis can be rejected.
1
Q
Define hypothesis.
A
A statement that an underlying tendency of scientific interest takes a particular quantitive value.
2
Q
What is the equation for the error factor?
A
Prevalence and incidence EF = e^(2 x rt(1/d))
IRR EF = e^(2 x rt(1/d1 + 1/d2))
SMR EF = e^(2 x rt(1/O))
d = number of cases, O = number of observed cases.
3
Q
How do you calculate the 95% confidence intervals?
A
Lower = result (ie. I, P IRR or SMR) / error factor Higher = result x error factor
