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;
17 Cards in this Set
- Front
- Back
causality requires
|
-covariation
-directionality -control of other possible causes |
|
correlation coefficient
|
=r
demonstrates value of variablitiy |
|
range
|
-1.00 to +1.00
positive:as scores one 1 variable inc, scores on other variable inc negative:as scores on 1 variable inc, scores on other dec |
|
magnitude
|
strength of linear relationship
farther from zero, the stronger the relationship (-.70 has great magnitude than .40) |
|
interpreting r
|
<0.10 is trivial
0.10< r <0.30 is weak 0.30< r <0.50 is moderate r>0.5 is strong r^2 tells how much variance is accounted for |
|
p-value
|
tells whether correlation is statistically significant or not
Null hypothesis: no relationship, r=0 alternative hypothesis: positive, r>0.00 or negative, r<0.00 |
|
probablity
|
if prob. that correlation = 0.00 is low (less than 5%) than can say correlation is statistically significant and reject null hypothesis
however, cannot 'prove' alternative hypothesis |
|
outliers
|
+/- 3 sd from the mean
can have big impact on r |
|
on-line vs. off-line
|
on-line can boost size of correlation
off-line can drag correlation down |
|
regression
|
prediciting scores on 1 variable based on the participant's scores on another variable
assess relationships of >1 predictors simultaneously |
|
multiple regression
|
assessing the relationship b/w mulitple predictor variables and an outcome variable
predicting an outcome variable based on predictor scores linear relationship, best fitting line |
|
moderator variables
|
interactions:when the relationships b/w a predictor and outcome variable depends on the value of another variable (moderator)
ex. licorice color; more ppl like red than black |
|
moderator variables can help understand...
|
can be a 3rd variable that influences the relationship b/w a predictor and outcome variable
understand the relationship establish boundary conditions |
|
types of moderator variables
|
Discrete: examine predictor-outcome variable for each category (licorice either black or red)
Continuous: one solution- change continuous to discrete high category- group > +1 sd low category- group < -1 sd |
|
moderator analysis
|
predictor, moderator, and new interaction variable (PxM) are entered into regression equation
high values: both P and M are large low values: either P, M, or both values are small |
|
mediator variables
|
an intervening 3rd variable that (partly) explains the influence of the predictor on the outcome
predictor incluences mediator, which influences outcome variable ex. viewing violent tv-->arousal (mediated effect)-->agressive play |
|
mediated effect if..
|
the magnitude of direct relationship b/w the predictor and outcome is drastically reduced (or 0) when the relationship b/w the mediator and outcome is considered
the predictor and mediator are related |