Term
What is standard deviation? |
|
Definition
| How much the scores deviate from the mean |
|
|
Term
What is an extreme score? |
|
Definition
| It is a score that is more than 2 std deviations away from the mean. |
|
|
Term
What are the chances of extreme scores in a data sets with a normal distribution?
|
|
Definition
| Extreme scores with more than 2 standard deviation from the mean will occur only 5% of the time. |
|
|
Term
How does 5% of extreme score principle apply to a normal distributed bell shape curve? |
|
Definition
| There is a 2.5% chance of extreme scores occuring on each tail/end of the curve. |
|
|
Term
| What is the general possibility of inference error (false alarm error rate/alpha level) allowed in Psychology? |
|
Definition
| Alpha criterion is .05. We give ourselves the 5% chance that we may have made a wrong inference from that data analysis. (eg. 5% Falsely rejecting the Null Hypothesis) |
|
|
Term
|
Definition
| It is an analysis of variance which compares the variation within groups to the variation between groups. |
|
|
Term
How does the null hypothesis relate to ANOVA and a hypothetical Psychology experiment? How does this relate to p < .05? |
|
Definition
Null Hypo is that the experiment had no effect, and that the differences/effect are due to chance to alone.
Thus p < .05 means that if null hypo is true, there is a less than 5% chance of observing differences due to chance found between samples. Hence null hypothesis is rejected.
|
|
|
Term
| What is the F statistics? How does it relate to signficance testing for differences not due to chance alone? |
|
Definition
| The F stat is a ratio of the variance of between group to variance of within groups. If a F stat is significantly large with a P < .05, there is significant difference not due to chance alone. |
|
|
Term
| What is SS-Within? What is SS-Between? What is formula for calulating F- stat? |
|
Definition
|
|
Term
| What are other tests are required after a significant difference from the one way anova? |
|
Definition
T-test= to determine where the difference between 2 group lies.
Bonferroni test= to do multiple comparisons to see where the differences lies and by how much. (For comparision of more than 2 groups)
Bonferroni also adjust p value to ensure there is a minimised chance of making wrong inferences which may occur in multiple comparisons. |
|
|
