Ap stats: chapter 9 Flashcards
in order to decrease Type II
Increase Type I
Use equal to and not equal to when
figuring out is something is fair/unfair (or when finding difference)
Interpret power
If the true ___ is really ___, then we have a __% chance of detecting a __.
significance level and confidence level MUST
MATCH (=100) in order to use them in the same situation
To increase power
increase n OR increase significance level OR increase type II
how to calculate outliers
IQR: Q3-Q1
low: Q1 - 1.5(IQR)
high: Q3 + 1.5(IQR)
when you fail to reject Ho its a type
II error
when you reject Ho its a type
I error
if the power is 80% then type II is ?
20% (them two equal 100)
To DECREASE type 11
INCREASE (choose the highest) significance level
when giving significance level and confidence level ALWAYS make sure to
compare / MATCH them (talk about them!)
test –> 5
one sample z-test for proportions
test statistic for proportions
z = p-hat - p / sqr-root p(q)/n
test –> 2
One sample t-test for Mu
AND
matched pair t-test for Mu
test statistic for Mu
t = mean - Mu / Sx / sqr-root n
if p-value is less than significance level
- reject Ho
- stat sig
DO
if p-value is more than significance level
- fail to reject Ho
- not stat sig
DO NOT
MATCHED PAIRS
- ## POI is the true average DIFFERENCE between ___.Ho: Mu = 0
Ha: Mu <,>,(not equal to) 0
—–
MATCHED PAIR t-test for Mu:
CLT check
n greater than or equal to 30
Null Hypothesis(Ho)
NO CHANGE
Alternative Hypothesis (Ha)
IS CHANGE
Interpret p-value
We do not have convincing evidence that the ____is less than/more than/different than ___.
PANIC conclusion
we are __% confident that the interval ___ to ___ capture the true proportion/average of ___.
interpret confidence level
if we make many __% confidence intervals, we expect about __% to capture the true mean/proportion.
Type I error occurs when
We say Ha, but it’s Ho
Type II error occurs when
We say Ho, but it’s Ha
Null is…
ASK ABOUT NULL
