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;
19 Cards in this Set
- Front
- Back
There are two ways to Test Ho when Ha is _ |
Two sided =- |
|
Using a _ and designating _ |
Significance test, Alpha(Fishy) |
|
If the P value _, then the data is statistically _ reject Ho. |
<- Alpha, Significant |
|
Using a _ and choosing your _ |
Confidence interval, C-confident |
|
If _ is outside the confidence interval, then the data is statistically significant and we _ |
Reject Ho with however much cont. |
|
In either case you will reject or fail to reject Ho exactly same as long as you're testing the two sided case (=-). In both cases, you decide the significance level _ or the confidence level _. |
Alpha, C |
|
Levels of Significance(How much evidence is required to reject Ho? (No answer for this one |
N/A |
|
P-Value Less than .01 |
Interpretation: Highly statistically significant Very evidence against the null |
|
P-Value .01 to .05 |
Interpretation: Statistically significant Adequate evidence against the null |
|
P-Value Greater that .05 |
Interpretation: Insufficient evidence against the null |
|
When do we need real strong evidence? |
Medical profession when $ is involved |
|
Use and Abuse of Tests |
A large sample will declare a very small difference statistically significant. |
|
Use and Abuse of Tests 2 |
Always look at shape of the data, if available. Be aware of outliers or other deviations in the pattern. |
|
Use and Abuse of Tests 3 |
Can make data significant or ruin good data always graph. |
|
Statistical Inference- not for all sets of data Studies of English vocabulary scores between students who take a foreign language and those you don't are often not conclusive. Why? |
College kids that take a foreign lang. |
|
How could you solve the problem? |
Set up a comparative study. |
|
Beware of the search for significance If you perform a lot of tests on one set of data, _ a few will appear statistically significant. Important |
By chance alone. Important |
|
Ex) Psychologists measured 77 variables on schizophrenic persons and non- schizophrenic people. They found 2 variables to be significant. Why is this not surprising? How many would you expect by chance alone |
.05 Times 77= 3.85 it 5% by chance. |
|
Don't interpret the P-value as the probability _ |
Ho is true (Prob of finding evid. like you did if Ho is true) |