Epidemiology and Stats Flashcards
Properties of the distribution of sample means
Mean of all sample means will be equal to population mean
Standard deviation of all sample means is called the Standard Error
Given a large enough sample size the distribution of sample means will always be normal
Equation for Standard Error of a study sample
SE = SD/root(N)
SD is standard deviation of sample
N is size of sample
Properties of Standard Error
Large means measurement is imprecise
Small means estimate is precise
As sample size increases SE gets smaller
Confidence Interval
Range of values in which the population parameter is likely to lie
Calculation of 95% CI
X+- 1.96xSE
Where X is the sample parameter (like mean0
Lack of significance in CI
If a CI for a difference includes zero it is insignificant
If a CI for a ratio includes 1 it is insignificant
Type 1 Error
Rejecting the null hypothesis when it is true
(Thinking data is significant when P is greater than Alpha)
Significance level of Alpha is probability of making a type I error
P=.05, Prob of type 1 error is 5%
False Positive Rate
Type II Error
Inability to reject the null hypothesis even when the data is actually significant
Denoted by B
Power
1-B
The chance of obtaining a statistically significant p-value when the null hypothesis is truly false
False negative rate