Flashcards in biostats II Deck (13):
what is standard error of the mean?
estimation of how much variability exists btw the sample mean and the true population mean
type I error
stating that there is an effect or difference when, in reality, none exists (finding that the alternate hypothesis, when the null hypothesis is true). aka false-positive error. (you sAw a difference that doesn't exist)
What is alpha?
the probability of making a type I error. p is judged against a pre-set measure of significance
type II error
stating that there is not an effect or difference when one exists (false-negative error; Blind to the difference that exists)
What is beta?
the probability of making a type II error. beta is related to statistical power (1-beta). power is the probability of rejecting the null hypothesis when it is false.
how do you incr. power/ decr. beta of a study?
incr. sample size, incr. expected effect size, or incr. precision of measurement. `
checks the differences btw the means of 2 different groups. (BP in men vs. women)
checks the differences btw the means of 3 or more groups (mean BP of 3 different ethnic groups)
checks the difference btw 2 or more precentages or proportions of categorical outcomes. (comparing the percentage of men vs. women with HTN)
pearson correlation coefficient
r is btw -1 and 1. the closer the absolute value is to 1, the stronger the linear correlation btw 2 variables. usually reported as r squared (coefficient of determination)p
positive vs. negative skew
positive skew: asymmytry with longer tail at the right (this is where the graph gets skewed). mean> median > mode
negative skew: meand < median < mode. asymmetry with longer tail at the left
how does resistance to aminoglycosides occur?
chemical modification of the drug