Core Knowledge in Scholarly Activities Flashcards
Test with high sensitivity
Patient with disease will have a positive test
Test with high specificty
Healthy patient will have negative test
The positive or negative predictive value of any test depends on this measurement
Prevalence of the condition in the population being tested
Correlation coefficient interpretation
1 = strong positive association
0 = no association
-1 = strong negative association
When to use a t test
Comparing two groups with continuous variables (e.g., IQ score)
When to use a paired t test
To compare the same patients (i.e., before and after)
When to use a Fisher exact test or chi-squared test
Comparing two groups with categorical variables (e.g., insured or uninsured)
Type I error in hypothesis tseting
False positive; p value = chance of type I error
Type II error in hypothesis testing
False negative; prevent it with a power calculation
What percentage of data on a normal curve falls within 2 standard deviations (SD)?
95%
(68% within 1 SD, 99.7% within 3 SD)