Week 9 Flashcards
What is the difference between systematic error and random error?
Random error is due to chance such as statistical fluctuations in the measurement of data. Systematic error is a result from errors in the selection or measurement such as bias.
What is the difference between internal validity and external validity?
Internal validity is the validity of inferences drawn as they pertain to the members of the source population. External validity or generalizability is the validity of the inferences as they pertain to people outside of the source population.
When does selection bias occur in the natural history of the study?
After the choosing the population and before selecting the study participants.
When does information bias occur in the natural history of a study?
During the data collection step.
Give an example of information bias
misclassification of former smokers as non-smokers
What is non-differential misclassification?
When the misclassification of exposure status, for instance, is equal in the diseased and non-diseased groups (ie, it does not differ by disease).
How does non-differential misclassification, which is independent of disease status, affect the estimates in a study?
non-differential misclassification biases toward the null (no effect)
Twenty percent of the nonusers with CHD were categorized as users. Everyone else was classified accurately according to the truth. Is this an example of misclassification of exposure or disease?
misclassification of exposure
Twenty percent of the nonusers with CHD were categorized as users. Everyone else was classified accurately according to the Truth. What type of misclassification is this an example of?
Differential misclassification of exposure, because the misclassification of exposure status is dependent on disease status.
How can differential misclassification affect the estimates of study?
Differential misclassification can bias either toward the null or away from the nul.
To have internal validity, a study must successfully address selection bias, information bias, and confounding, but not
generalizability
In an observation study, threats to external validity include
effect modification
What is validity?
The degree to which the inferences drawn from a study are warranted when account is taken of: the study methods, the representativeness of the study sample, and the nature of the population from which it is drawn.
What is internal validity?
the credibility with which statements can be made about whether the association between the exposure and disease found in your study is a true association; what you observe is not due to random error, confounding or bias
What is external validity (generalizability)?
validity of the inferences as they pertain to other circumstances; ability to generalize the results from a given study to populations beyond the study subjects.
What study design is limited in generalizability?
clinical design, experimental study design; however, the bigger the sample, the more it is generalizable
What are some threats to internal validity?
chance (random error), confounding, bias
What are some threats to external validity?
internal validity, choice of study population, effect modification