U7 Flashcards
what is a correlation
- not cause and effect
- describes relationship between 2 variables
correlations indicate …
and are described by _____ and _____
- degree of consistency and variability between 2 variables
direction and strength
regression analysis and what it indicates
- how closely scatterplot fits regression line (line of best fit)
- indicates relative accuracy of a prediction
the ______ the relationship between variables is the ____ the accuracy when trying to predict
stronger
better
r value of perfect correlation and no correlation
perfect positive = 1.00
perfect negative = -1.00
none = 0.00
PPMC (Pearson product-moment correlation)
- r value
- determines strength between 2 variables in a linear relationship
- relies on covariance (how x and y vary together)
- greater r more predictive of relationship
low correlation doesn’t eliminate ________ nor does a high correlation prove it
causation
little to no relationship
fair relationship
moderate to good relationship
good to excellent relationship
= 0.00 - 0.25
= .25 - .50
= .50 - .75
= .75 +
strength of relationships in social sciences
0.00 - 0.10 = weak
0.11 - 0.30 = moderate
0.30 + strong
inverse vs direct linear relationships
inverse = decline
direct = incline
limitations of PPMC
- 2 groups w no relationship can produce a strong correlationand vice versa
- extreme data points can have large effect
- relationship could be present but not evident from PPMC since not linear relationship
coefficient of determination
how much one variable can explain the degree of variation in a pint from the regression line
equation of a straight line
Y^ = a +bx
Y^ = predicted y value
x = measured x value
a = y intercept
b = slope regression line
residual
(Y-Y^)
measures magnitude of error of prediction
the predicted value of a score won’t = raw scores unless r =
1.00
SEE
sd of an average
for predicting future data (avg score for x-value +/- value)
RCT = randomized control trial
establishes cause and effect (efficiency)
- placebo and test group
- random assignment
t or f
examinning correlations between previous data can prove cause and effect relationships
f
negative r value = ________ relationship
positive r value = ______ relationship
inverse
direct
null hypothesis vs alternative hypothesis
H0: p=0
H1: p not = 0
test statistic must be _______ then or equal to critical value to reject H0 (signifying a significant relationship)
greater
