Term
| experimental research vs correlational research* |
|
Definition
-experimental - manipulates variables and observes outcome, focus on general laws
-correlational - observe variables and relate them, focus on individual differences |
|
|
Term
|
Definition
| high scores on one variable associated with high scores on a second V |
|
|
Term
|
Definition
| high scores on one V associated with low score son a second V |
|
|
Term
|
Definition
-assumes linear relationship
-avoid restricting the range (weakens correlation therefore decreases ability to make predictions |
|
|
Term
| coefficient of determination (eg r = 1.00, 0.5) |
|
Definition
-r^2
-proportion of variability in one variable that can be accounted for by variability in the other variable
-r^2 = 1.00 --> 100% of variability in VI accounted for by V2
-r^2 = 0.25 --> 25% of variance in V1 accounted for by V2 (75% by other factors) |
|
|
Term
|
Definition
-making predictions for Y, knowing X and correlation size
-use regression line |
|
|
Term
|
Definition
straight line that best summarizes a correlation
Y = a + bX
Y: criterion variable (V being predicted)
X: predictor variable (V making prediction) |
|
|
Term
| probelms in interpreting correlations and their solution (2) |
|
Definition
1. directionality problem
soltn: cross laggd panel correlation (measure relationship between two main variables, separated in time)
2. 3rd variable problem
soltn: partial correlation (control for 3rd variables statistically by "partialing out" - post facto equivalent groups) |
|
|
Term
| correlation and psychological tests, deciding reliability/validity (3) |
|
Definition
1. split half reliability - 1/2 items of a subtest, correlate with other 1/2
2. test-retest reliability - correlation between two test administrations
3. criterion validity - correlation of test scores to scores on criterion |
|
|
Term
| multiple regression: formula |
|
Definition
-making predictions for multivariate correlations
Y = a + b1X1 + b2X2..
Y: criterion variable (V being predicted)
X: predictor variables (Vs making prediction)
b: regression weights (determine relative importance of each X)
-advantage! combination of X influences increases prediction |
|
|
