Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
31 Cards in this Set
- Front
- Back
|
what measures exactly how unusual the results of an experiment really are
|
p-values
|
|
|
the probability of rejecting a null hypothesis that is false
|
power
|
|
|
power depends on what 4 things?
|
1. the degree to which the null hypothesis fails
2. significance level of the test 3. standard deviations 4. sample size |
|
|
probability to detect a given difference from hypothesized value
|
power
|
|
|
as type I error decreases, type II error...
|
increases
|
|
|
proability of rejecting the null hypotheis when it is true
|
type 1 error
|
|
|
proability of accepting the null hypotheis when it is false
|
type 2 error
|
|
|
as sample size increases, power
|
increases
|
|
|
as the difference between means increases, power
|
increases
|
|
|
as variance decreases, power
|
increases
|
|
|
Test the null hypotheis, H0: u1=u2
|
unpaired two-sample t-test
|
|
|
Tests that the means of some variable are equal in two groups of subjects
|
unpaired two-sample t-test
|
|
|
Assumptions of unpaired two-sample t-test
|
1. independent samples
2. normal distrubutions (in each group) 3. equal standard deviation |
|
|
distribution and degrees of freedom for unpaired two-sample t-test
|
t-distribution,DF= n1 + n2 - 2
|
|
|
as the number of degree of freedoms increase, the t-distribution becomes...
|
more normal
|
|
|
tests the H0: uD = 0
Test the null hypothesis that mean difference between of some variable is 0 |
paired (one sample) t-test
|
|
|
distrubtion and degree of freedome for paired t-test
|
t-distribution
DF= n-1 |
|
|
assumptions of paired t-test
|
normal distribution
independent sample |
|
|
Wilcoxon's test is an alternative to what test?
|
paired t-test
|
|
|
when do you use the wilcoxon's test
|
when the assumptions for the t-test are not satisfied
|
|
|
advantages of wilcoxon's test
|
1. only required assumption is tha the samples are indepdent
2.robust! (outliers have little effect on results) |
|
|
disadvantage of wilcoxon's test
|
1. less powerful than t-test
2. confidence bounds not easily computed |
|
|
which test of means is more powerful? t-test or wilcoxon's
|
t-test
|
|
|
assumptions for wilcoxon's test
|
indepdendent samples
|
|
|
when testing means, what are the parametric tests
|
unpaired t-testpaired t-test
|
|
|
when testing means, what is the non-parametric test
|
wilcoxon's
|
|
|
Tests the mean differences between more than
two groups |
ANOVA
|
|
|
HO: u1=u2=u3
|
ANOVA
|
|
|
pros of ANOVA
|
comparing more than 2 tests simultaneously
|
|
|
cons of ANOVA
|
diminishing returns! adding a factor adds a degree of freedom which decreases power
|
|
|
the ANOVA is comprabile to what test
|
unpaired t-tests
|
