Confidence intervals, p-values + hypothesis testing Flashcards
What is the standard deviation?
Describes variability of observations in the sample
What is the standard error
Measure of the precision of an estimate of the population paramter
What is variance?
The value of deviation from the mean
How is variance calculated?
(Σ(x−x(bar))^2)/(n−1)
The sum of the difference between values and the mean, squared, divided by n-1
How is standard deviation calculated?
Square root of variance
How is standard error of the mean calculated?
SD/√n
Standard deviation/root of sample size
How is standard error of a proportion calculated?
√(p(1-p))/n
Root of proportion divided by sample size
How is standard error of difference between two sample means calculated?
Used to compare two independent samples
√(SD^2/n)+(SD^2/n)
Sample standard deviation squared/sample size aded together
How is standard error of the difference between two proportions calculated?
√(proportion 1)+(proportion 2)
Proportion = p(1-p)/n
How is the 95% confidence interval calculated for the mean?
x(bar)+/-[1.96xSE(mean)]
How is the 95% confidence interval of a proportion calculated?
p+/-[1.96xSE(proportion)]
How is the 95% confidence interval calculated for the difference in two means?
x(bar1)-x(bar2)+/-[1.96xSE(difference in means)]
How is the 95% confidence interval calculated for the difference in two proportions?
p1-p2+/-[1.96xSE(difference in proportions)]
What does the p value tell us?
The probability of obtaining these results, or more extreme, when the null hypothesis is true
What is a type I error?
False positive
Rejecting the null hypothesis when it is true
Given as α
What is a type II error?
False negative
Not rejecting the null hypothesis when it is actually false
Given as β
What is the power?
Probability of rejecting the null hypothesis when it is actually false
How is power calculated?
1-β
What does a 95% confidence interval for a mean show us?
If the study were repeated 100 times, of the resulting 100 CI for the mean, we would expect 95 of these to include the true population mean
It is a range of values which will include the true population mean for 95% of possible samples
How is the test statistic (Z) calculated?
(Observed value-hypothesised value)/Standard error of observed value
Usually hypothesised value = 0
