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;
52 Cards in this Set
- Front
- Back
What is the equation to determine the chance that we will always be right? |
(1-a)^n |
|
What is the equation to determine that chance of having 1/more Type 1 Errors? |
1-(1-a)^n |
|
________ alpha is BIGGER than our _______ alpha |
Experiment-wise; Test-wise |
|
What is the formula for the Bonferronni Correction? |
Atest = Aexperiment/n |
|
How does the Bonferroni Correction do? |
makes the test-wise alpha smaller |
|
What might make 2 participants score differently on DV if they are in the SAME condition? |
Random effects; chance |
|
What might make 2 participants score differently on DV id they are in DIFFERENT conditions? (3) |
1. Effects of Treatment 2. Random Effects; chance 3. Effects of Confounds |
|
If the null hypothesis is true, how likely is it that a difference this big (or bigger) would occur? |
the p value |
|
If the p value is really SMALL (p<.05), then you would __________ the null hypothesis? |
REJECT |
|
If the p value is BIGGER (p>.05), then you would _________ the null hypothesis? |
FAIL TO REJECT |
|
The P value must be smaller than the Alpha level for us to say that it is _______________ |
statistically significant |
|
What does the Alpha level mean? |
the chance of making a Type 1 Error |
|
What is a Type 1 Error? |
Wrongly Rejecting the Null Hypothesis |
|
What is a Type 2 Error? |
Accepting a False Null Hypothesis |
|
What are 4 things that affect the chance of a Type 2 Error? |
1. effect size 2. a-level 3. sample size 4. error variance |
|
What is the symbol for a Type 2 Error? |
β |
|
Why don't we know β? (2) |
Because we can only ESTIMATE: 1. Error Variance 2. TRUE effect size |
|
What is Power? |
likelihood that we will CORRECTLY reject the null when the null is false |
|
What is the equation to find the Power? |
Power = 1 - β |
|
Why would you do a Power Analysis? |
to determine how many participants we will need to detect an effect of a particular size |
|
What are 5 ways that you can increase Power? |
1. by increasing alpha 2. by increasing # of participants 3. by DECREASING error variance 4. by increasing experimental effect 5. by making β smaller |
|
What does a T-test do? |
tests the difference between 2 means |
|
People in the 2 independent samples T-test _______ have unique connections between them. |
DO NOT |
|
People in the Matched or Paired Samples T-tests _______ have unique connections between them. |
DO |
|
What are the drawbacks to T-tests? |
1. can only test 2 means at a time 2. requires repeated use in more complex designs |
|
The chance of making one or more Type 1 Errors ____________ with the number of T-tests |
INCREASES |
|
Formula for: 2 tests, always right |
(1-a) x (1-a) |
|
Formula for: wrong at least once |
1 - chance we will never be wrong |
|
Formula for: 3 tests, always right |
(1-a)^3 |
|
What is the advantage of the Bonferroni Correction? |
protects against Expriments-wise Type 1 Error inflation |
|
What is the drawback of the Bonferroni Correction? |
decreases a-test, thereby INCREASING likelihood of Type 2 Error |
|
The likelihood of Type 2 Error _________ with the number of tests |
INCREASES |
|
What is ANOVA? |
an inferential statistic |
|
What does ANOVA allow us to do? |
can make a reasonable "inference", meaning that we can make a decision with a degree on confidence |
|
For Between-subjects Designs, ANOVA compares ______________ to ______________ |
Systematic Variance to Error Variance |
|
What is the F statistic? |
MSsystematic / MSerror |
|
What is MSsystematic? |
Mean Sqared Devation - Systematic |
|
What does MSsystematic do? |
estimates the average amount of variance associated with manipulation of the IV. |
|
What is MSerror? |
Mean Squared Deviation -- Error |
|
What does MSerror do? |
estimates the average amount of variance due to error or chance |
|
The bigger F gets, the ___________ that MSsystematic gets in comparison to MSerror. |
BIGGER |
|
When F gets bigger, we get to _________ the null hypothesis |
reject |
|
What are Total Degrees of Freedom? |
# of scores that would vary freely until we would know for certain exactly what the remaining scores would have to be in order to get the calculated value f the Grand Mean |
|
What is the equation for the Degrees of Freedom? |
df = n-1 |
|
How can you calculate the Degrees of Freedom of individual scores from within groups? |
dfwg = # of participants - # in groups |
|
Degrees of Freedom represent the # of __________ for a SS to INCREASE |
opportunities |
|
How would you find the Mean Square Within Groups? |
MSwg = SSwg / DFwg |
|
If Fobs > Fcrit then we would _________ the null |
REJECT |
|
If Fobs < Fcrit, then we would __________ the null |
RETAIN |
|
What are the 3 F tests conducted for a 2-way Between Subject Design? |
1. for main effect of Factor A 2. for main effect of Factor B 3. for main effect of Factor A with Factor B |
|
How many test should you do? For 3 means? For 4 means? |
3 tests (1&2, 1&3, 2&3) 6 tests (1&2, 1&3, 1&4, 2&3, 2&4, 2&4) |
|
What do Post hoc Tests control for? |
Type 1 Error |