exam 2 Flashcards
how are hypotheses tested 4 steps
- State the hypothesis
- compute test statistic
- determine p value
- draw conclusion
what is the null distribution
the probability distribution of test statistic assuming null hypothesis is true. If observed falls within extremes we reject the null.
if P<0.05 then
we reject the null hypothesis. there is an association
if P >0.05 then
we fail to reject the null hypothesis- accept
what is type one error
rejecting a true null hypothesis, false positive
why is type 2 error
failing to reject a false null hypothesis, false negative
what is power
correctly rejecting the null hypothesis when it is actually false
if there is low error there is high what
power
how do you get high power but low type 2 error
large samples, large true discrepancy, low variability in population, larger significance level
if 10<15<20 what is the conclusion
fail to reject null hypothesis
how do you calculate number of successes in a random sample
probability x n
how do you estimate a proportion
x/n
what are the rules for a goodness of fit test
- none of the categories have an expected frequency of <1
- no more than 205 of the categories have expected frequencies <5
if they do- combine
what are critical values
marks the boundary of a specified area in the sampling distribution under Ho
pick the one at df and 0.05 and see if p value is greater or lower
if s^2 > mean what would the distribution look like
clumped
2 categorical variables not associated implies what
the 2 variables are not associated- independent
if the confidence interval excludes 1 in odds odds ratios
the association is statistically significant
the significance level is what
threshold for statistical significance
what are ways to decrease type 1 error
decrease statistical significance
ways to reduce type 2 error
increase sample size, increase statistical significance ( increases tyoe 1) , use more sensitive test
OR > 1 means
OR > 1 → Exposure increases risk.
