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;
8 Cards in this Set
- Front
- Back
how to calculate the false positive ratio?
|
1 - specificity
|
|
how to calculate NNT?
|
1 / ARR
ARR = Experimental Event Rate - Control ER |
|
how is power of a study defined?
|
probablity of a difference when there is one
power = 1 - beta (beta = type II error FINDING NO difference when one ACTUALLY exists) |
|
what are chi-squared test, t-test, and ANOVA test for?
|
chi-squared - comparing PROPORTIONS, how many fall into each category
t-test - comparing MEAN values ANOVA - comparing means of 3+ variables |
|
how is the median calculated when there are an even # of data points?
|
take the center 2 points and the median is halfway between them
|
|
how to calculate the likelihood ratio?
|
LR + = sensitivity / (1-specificity)
> 10 significantly increases likelihood pt will have dz for which tested positive |
|
how to calculate confidence intervals?
|
mean, SD, Z-score, sample size
first, calculate standard error of the mean (SEM) = SD / square root (n) multiply SEM by z-score z-score = 1.96 for CI 95%, = 2.58 for CI 99% so, for 95% CI = 1.96*SD/square root (n) |
|
what is Berkson's effect? Pygmalion? Hawthorne?
|
Berkson's bias - selection bias created by selecting hospitalized patients as control
Pygmalion effect - researches beliefs in efficacy of treatment that can potentially affect outcome Hawthorne effect - tendency of a study population to affect an outcome due to knowledge being studied |