Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
9 Cards in this Set
- Front
- Back
sensitivity
|
percent who test positive
among all who have a disease (1 - sensitivity) are false negatives, so high sensitivity rules out (SNOUT) |
|
specificity
|
percentage who test negative
among all who don't have a disease (1 - specificity) are false positives, so high specificity rules in (SPIN) |
|
positive predictive value
|
percent of positive test results
that actually represent disease |
|
negative predictive value
|
percent of negative test results
that actually represent no disease |
|
odds ratio
|
from case-control (retrospective) study comparing disease to prior risks
( WITH disease WITH risk x WITHOUT disease WITHOUT risk ) ÷ ( WITH disease WITHOUT risk x WITHOUT disease WITH risk ) approximates relative risk for dz with low prevalence |
|
relative risk
|
from cohort (prospective) study
risk of getting disease (D) based on risk (R) ( WITH disease / ALL WITH risk ) ÷ ( WITH disease / ALL WITHOUT risk ) |
|
type I error
|
stating there IS an effect when NONE exists;
to incorrectly reject null hypothesis p represents odds of type I error |
|
type II error
|
stating there is NO effect when one EXISTS;
to incorrectly accept null hypothesis |
|
power
|
probability of finding an effect (rejecting null hypothesis) when the effect is real
(+) n -> (+) power power = ( 1 - type II error ) |