Distinguish between reproducibility and validity.
Distinguish between a sign and symptom. Is nausea a sign or symptom? Is vomiting a sign or symptom? Is a red throat a sign or symptom? Is a sore throat a sign or a symptom?
What statistics is best for quantifying the reproducibility of a diagnosis?
What is the advantage of using the kappa statistic instead of "percent agreement"?
True or false? A kappa statistic of 0 indicates no agreement. Explain your response.
Define: (a) true positive (b) true negative (c) false positive (d) false negative.
Match the statistic with the conditional probability it estimates. Statistic: SEN, SPEC, PVP, PVN Conditional probability : (a) Pr(T-| D-) (b) Pr(T+|D+) (c) Pr(D+|T+) (d) Pr(D-|T-)
This is the proportion of disease-free individuals will have a negative test.
This is the proportion of negative tests that will indicate the person is disease-free.
What factors determine the PVP of a test?
Why is PVP typically low when prevalence is low?
Consider a test in which high values indicate presence of a condition. Assume there is overlap in test scores between diseased and non-diseased populations. What effect will decreasing the cut-off point for a positive test have on the sensitivity of the test?
Should the first stage of a screening program that aims to detect all cases opt for high sensitive or highly specific? Explain.
Key Last update: 05/06/2009