Home
Subjects
Explanations
Create
Study sets, textbooks, questions
Log in
Sign up
Upgrade to remove ads
Only $2.99/month
Correlation & Regression Exam 2
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (37)
what correlation coefficient to use for two interval/ratio scales
Pearson's r
what correlation coefficient to use for tied ordinal scales
Pearson's r
what correlation coefficient to use for anything involving an ordinal scale
Spearman's r
what correlation coefficient to use for two true dichotomies
phi
what correlation coefficient to use for one interval/ratio scale and one true dichotomy
point biserial
what correlation coefficient to use for one interval/ratio scale and one dichotomized variable
biserial
what correlation coefficient to use for two dichotomized variables
tetrachoric
parameter coefficient symbol for Pearson's r
rho (ρ)
parameter coefficient symbol for Spearman's r
rho-sub-s (ρs)
parameter coefficient symbol for phi
ϕ
parameter coefficient symbol for point biserial
rho-sub-pb (ρpb)
parameter coefficient symbol for biserial
rho-sub-bis (ρbis)
when to use Spearman's r instead of Pearson's r
(1) both are ordinal; (2) data isn't normally distributed
what happens when you dichotomize a variable
(1) lose information; (2) lower magnitude of resulting correlation (lower than Pearson's r)
what "f(Z)" represents
height of the normal distribution curve where Z cuts off p and q
maximum value of phi
not always 1--only 1 if the marginals are equal
relationship between biserial & point biserial in terms of value
biserial is always stronger because it uses more data
standard regression equation
Zy (hat) = β1 * Zx
β1 in a standardized equation
r
when the best fit line has a slope equal to r
when standard deviations are equal
how to describe the interpretation of the standard regression weight
For every standard deviation increase in Y, there is an X standard deviation increase in X
coefficient of determination
proportion of variance due to the regression
coefficient of nondetermination/alienation
proportion of variance due to error
k
number of predictors (linear regression: 1)
symbol for the regression coefficient
b1
partial correlation
correlation between X1 and Y controlling for the relationship between X2 and X1 and Y (both)
semipartial correlation
correlation between X1 and Y controlling for the relationship between X2 and X1
symbol for the variance of Y due to X1
r-sub-Y1
symbol for the variance of Y due to X2
r-sub-Y2
symbol for the variance of Y due to X1 & X2
R-sub-Y.12 (multiple correlation)
symbol for the relationship between two variables Y & X1 when variability due to X2 is removed from X1
r-sub-Y(1.2) (semipartial correlation)
symbol for the variance between X1 & X2
r-sub-12
relationship between two variables Y & X1 when variability due to X2 is removed from X1 and Y
r-sub-Y1.2 (partial correlation)
meaning of r-sub-Y1
variance of Y due to X1
meaning of r-sub-Y2
variance of Y due to X2
meaning of r-sub-12
variance between X1 and X2
symbol for the standardized regression weight
β1
Sets with similar terms
RMS Final
40 terms
Statistics: Correlation 1
49 terms
Psy 360 Statistics USM
35 terms
Chapter 8 Review One
28 terms
Other sets by this creator
My EPPP Study Cards
76 terms
Ethics Exam
46 terms
Psychometrics Exam 4
55 terms
Psychometrics Exam 1
69 terms