SCS Study Guide- EBP Flashcards
if a test is positive and they really have the disease, it is a ____-
true positive
if the test is positive and they do not have the disease, this is a
false positive
if the test is negative, but they have the disease
false positive
if the test is negative and they do not have the disease
true negative
sensitivity
those with the disease who have a positive result
looking at true (+)
SNOUT
specificity
portion of those without the disease that have a true (-)
true (-) rate,
SPIN
sensitivity avoids the
false negative
specificity avoids the
false positive
type I error
false (+), using a treatment that is no better than the alternative
type II error
false (-)
not using a treatment because it said no difference, but in reality, it was better
if you have a high sensitivity, the rate of having a false (-) must be…
low
if you have a high specificity, the rate of having a false (+) must be…
low
positive predictive value
proportion of positive tests that are true positives
negative predictive value
those negative tests that are true negatives
as prevalence increase, what happens to PPV and NPV
PPV increases and NPV decreases
is LR affected by prevlanace
no, minimally
face validity
the instrument tested what is was supposed to test
content validity
sample measurements represent the whole
criterion related
concurrent and predictive
relationship to gold standard.
concurrent: validity when measures are taken at the same time
predictive: outcome can be used to predict future outcomes
construct
assesses ability of instrument to measure a theoretical hypothesis
case study
looks at one thing at depth
case-control study
descriptive research, subjects are chosen based on having/no-having disease, and then determines a previous exposure
longitudinal study
collect data over time, usually looking to describe developmental changes
randomized control trial
clinical treatment compared to control, subjects randomly assigned.
alpha levels
level of stat sig
risk of type I error
beta
type II error
ANOVA
comparison of 3 or more treatment groups
level or significance
probability that a finding could be due to chance, traditionally set at 0.05
dependent variable
variable response depends on another
independent variable
variable manipulated or controlled by investigator
Hawthorne effect
subjects knowledge of being part of study effects performance
interrater reliability
degree to which two or more testers can obtain same rating on multiple occasions
intrarater reliability
degree to which one tester can obtain same findings on multiple occasions
kappa
0.81-1: almost perfect agreement
0.61-0.8: substantial agreement
standard error of measurement
estimating standard error in set of repeated scores
minimal clinical importance difference
amount of change which represents actual change to a patient