Correlational Studies Flashcards
True or False
This is an example of a chi square test; 2 anesthetics are compared, where the individual only receives one injection (data is presented in a “contingency table”
True
You can obtain the expected values of the data in a contingency table by taking the ________ of numbers in the cells and multiplying it by the ______ number
probability
total
Once we have the observed and expected frequencies for each cell in a contingency test, we can look at the __________ and see if it could have occurred by ______
difference
chance
True or False
A contingency test is a family of curves depending on the sample size/number of variables
False; chi-square test
When calculating ____ _____ you take the observed value - expected value and square, then divide by the expected
chi square
The bigger the number of chi square the ______ the difference is
bigger
Chi-square assumptions include, both groups are _______, composed of _______ or _____ data, the categories are _______ exclusive (either a yes or no), and all expected frequencies are greater than or equal to _____
independent nominal ordinal mutually 5
________ refers to how consistent or reproducible observations or measurements are
reliability
True or False
An example of intra-reliability is two different examiners on the same day
False; that is an example of inter-relaibility
intra-reliability is where it is one examiner on different days
Reliability coefficients include ____-____ correlation and, ______ and _____ kappa statistics
intra-class
weighted
simple
Weighted kappa statistics involve _____ data (i.e. slightly off/way off/progressive scale)
ordinal
Simple kappa statistics involve _____ data (i.e. yes or no)
nominal
Intra-class correlation involves _____ or _____ data (i.e. set scale; 1, 2, 3, 4 and must choose a point)
interval or ratio
A reliability coefficient of >_____ is considered excellent, and <____ is considered poor
- 75
0. 40
_______ refers to how closely your, observations, measurement instrument, or measurements are to the actual or accepted values ( i.e. the “gold standard”)
validity
_______ tests indicate presence or absence of disease; they are usually compared to the “gold standard”
diagnostic
_______ answers the question of “if disease is present how often is the test positive”; it is proportional with the true diseases that is diagnosed as having disease by the test
sensitivity
If you have a high ________, then false negatives are low
sensitivity
_______ answers the question of “if disease is absent, how often is the test negative”; ability of test to correct identity those without disease; useful for ruling in presence of disease
specificity
If _______ is high, then false positives are low
specificity
Pneumonic for remembering sensitivity vs specificity;
SPIN
SNOUT
high specificity rules IN disease
high sensitivity rules OUT disease
Specificity and sensitivity are stable; and do _____ ____ depending on underlying disease, prevalence, and number of patients tested
not change
______ is the portion of results that agrees with the “gold standard”
accuracy