Lecture 4 Flashcards
1
Q
internal validity
A
- the degree to which the results obtained from a study reflect the truth
- truth as if everyone in source population participated
2
Q
source population
A
- population from which the sample was drawn
3
Q
external validity
A
- the degree to which the results of the study are applicable to other (external) populations that are not being sampled
- generalizability
4
Q
external validity in a clinical trial
A
- based on the degree to which the intervention itself can be replicated
5
Q
bias
A
- systematic error
- difference between the observed findings and the truth
- does not imply wicked intent
6
Q
the greater the bias
A
- the lower the internal validity
7
Q
magnitude of bias
A
- size of difference between observed effect and truth
8
Q
direction of bias based on
A
- relationship among
- truth
- observed findings
- null value
9
Q
bias away from null
A
- makes it look like a stronger association than it really is
- observed findings are farther from null value than the truth is from null value
10
Q
bias towards null
A
- makes it look like a weaker association than it really is
- observed findings are closer to null value than truth is to the null value
11
Q
confounding bias
A
- distortion in results due to the effects of a third, extraneous factor
- always a threat to the validity of observational studies
12
Q
how to prevent confounding bias
A
- randomization
- matching
- restriction
13
Q
selection bias
A
- distortion in results due a difference between subjects who participated in a study and those who did not
14
Q
how selection bias develops
A
- the way people are selected for the study
- refusal of eligibles to participate
- withdraw or loss to followup
15
Q
measurement bias
A
- distortion in the results due to error in measurement of exposure, outcome, or both
- easiest to understand by looking at yes/no
