45
Bivariate Linear Correlation

Bivariate Linear Correlation. Linear Function Y = a + bX

Embed Size (px)

Citation preview

Page 1: Bivariate Linear Correlation. Linear Function Y = a + bX

Bivariate Linear Correlation

Page 2: Bivariate Linear Correlation. Linear Function Y = a + bX

Linear Function

•Y = a + bX

Page 3: Bivariate Linear Correlation. Linear Function Y = a + bX

Fixed and Random Variables

• A FIXED variable is one for which you have every possible value of interest in your sample.– Example: Subject sex, female or male.

• A RANDOM variable is one where the sample values are randomly obtained from the population of values.– Example: Height of subject.

Page 4: Bivariate Linear Correlation. Linear Function Y = a + bX

Correlation & Regression

• If Y is random and X is fixed, the model is a regression model.

• If both Y and X are random, the model is a correlation model.

• Psychologists generally do not know this• They think

– Correlation = compute the corr coeff, r– Regression = find an equation to predict Y

from X

Page 5: Bivariate Linear Correlation. Linear Function Y = a + bX

Scatter Plot

Perfect Positive Linear

X

Y

Page 6: Bivariate Linear Correlation. Linear Function Y = a + bX

Perfect Negative Linear

X

Y

Page 7: Bivariate Linear Correlation. Linear Function Y = a + bX

Perfect Positive Monotonic

X

Y

Page 8: Bivariate Linear Correlation. Linear Function Y = a + bX

Perfect Negative Monotonic

X

Y

Page 9: Bivariate Linear Correlation. Linear Function Y = a + bX

Nonmonotonic Relationship

Test Anxiety

Per

form

ance

Page 10: Bivariate Linear Correlation. Linear Function Y = a + bX

For the data plotted below, the linear r = 0, but the quadratic r = 1.

Page 11: Bivariate Linear Correlation. Linear Function Y = a + bX

Burgers (X) and Beer (Y)

Subject X Y XY 1 5 8 40 2 4 10 40 3 3 4 12 4 2 6 12 5 1 2 2

Sum 15 30 106 Mean 3 6 St. Dev. 1.581 3.162

A Scatter Plot of Our Data

Burgers

Bee

rs

Page 12: Bivariate Linear Correlation. Linear Function Y = a + bX

Burger (X)-Beer (Y) CorrelationSubject X Y XY

1 5 8 40 2 4 10 40 3 3 4 12 4 2 6 12 5 1 2 2

Sum 15 30 106 Mean 3 6 St. Dev. 1.581 3.162

N

YXXYYYXXSSCP

))(())((

.

N

YYYYYYYYSSy

))(())((

165

)30(15106

Page 13: Bivariate Linear Correlation. Linear Function Y = a + bX

))(( YYXXSSCP

Page 14: Bivariate Linear Correlation. Linear Function Y = a + bX

))(( YYXXSSCP

Page 15: Bivariate Linear Correlation. Linear Function Y = a + bX

))(( YYXXSSCP

Page 16: Bivariate Linear Correlation. Linear Function Y = a + bX

Burger (X)-Beer (Y) CorrelationSubject X Y XY

1 5 8 40 2 4 10 40 3 3 4 12 4 2 6 12 5 1 2 2

Sum 15 30 106 Mean 3 6 St. Dev. 1.581 3.162

165

)30(15106

))(())((

N

YXXYYYXXSSCP

.

.44

16

1

N

SSCPCOV 80.

)162.3(581.1

4),(

yxss

YXCOVr

Page 17: Bivariate Linear Correlation. Linear Function Y = a + bX

Hø: ρ = 0

• df = n – 2 = 3• Now get an exact p value and construct a

confidence interval

309.264.1

38.

1

22

r

nrt

Page 18: Bivariate Linear Correlation. Linear Function Y = a + bX

Get Exact p Value

• COMPUTE p=2*CDF.T(t,df).

Page 19: Bivariate Linear Correlation. Linear Function Y = a + bX

Go To Vassar

• http://vassarstats.net/

Page 20: Bivariate Linear Correlation. Linear Function Y = a + bX
Page 21: Bivariate Linear Correlation. Linear Function Y = a + bX

N increased to 10.

Page 22: Bivariate Linear Correlation. Linear Function Y = a + bX

Presenting the Results

• The correlation between my friends’ burger consumption and their beer consumption fell short of statistical significance, r(n = 5) = .8, p = .10,95% CI [-.28, .99].

• Among my friends, beer consumption was positively, significantly related to burger consumption, r(n = 10) = .8, p = .006,95% CI [.34, .95].

Page 23: Bivariate Linear Correlation. Linear Function Y = a + bX

Assumptions

1. Homoscedasticity across Y|X2. Normality of Y|X3. Normality of Y ignoring X 4. Homoscedasticity across X|Y5. Normality of X|Y6. Normality of X ignoring Y• The first three should look familiar, we

made them with the pooled variances t.

Page 24: Bivariate Linear Correlation. Linear Function Y = a + bX

Bivariate Normal

Page 25: Bivariate Linear Correlation. Linear Function Y = a + bX

When Do Assumptions Apply?

• Only when employing t or F.• That is, obtaining a p value• or constructing a confidence interval.

Page 26: Bivariate Linear Correlation. Linear Function Y = a + bX

Shrunken r2

• This reduces the bias in estimation of • As sample size increases (n-1)/(n-2)

approaches 1, and the amount of correction is reduced.

52.3

)4)(64.1(1

)2(

)1)(1(1

2

n

nr

Page 27: Bivariate Linear Correlation. Linear Function Y = a + bX

Do not use Pearson r if the relationship is not linear. If it is monotonic, use Spearman rho.

Page 28: Bivariate Linear Correlation. Linear Function Y = a + bX

Every time X increases, Y decreases – accordingly we

have here a perfect, negative, monotonic relationship

Page 29: Bivariate Linear Correlation. Linear Function Y = a + bX
Page 30: Bivariate Linear Correlation. Linear Function Y = a + bX

Pearson r measures the strength of the linear relationship. Notice that it is NOT perfect here.

Page 31: Bivariate Linear Correlation. Linear Function Y = a + bX

Spearman rho measures the strength of monotonic relationship. Notice that it IS perfect here.

Page 32: Bivariate Linear Correlation. Linear Function Y = a + bX

Uses of Correlation Analysis

• Measure the degree of linear association• Correlation does imply causation

– Necessary but not sufficient– Third variable problems

• Reliability• Validity• Independent Samples t – point biserial r

– Y = a + b Group (Group is 0 or 1)

Page 33: Bivariate Linear Correlation. Linear Function Y = a + bX

Uses of Correlation Analysis

• Contingency tables -- Rows = a + bColumns

• Multiple correlation/regression

pp XbXbXbaY 2211

HighSchoolpMathVerbalECU GPAbSATbSATbaGPA 21

Page 34: Bivariate Linear Correlation. Linear Function Y = a + bX

Uses of Correlation Analysis

• Analysis of variance (ANOVA)

• PolitConserv = a + b1 Republican? + b2 Democrat?k = 3, the third group is all others

• Canonical correlation/regression

11?22?11 kk GroupbGroupbGroupbaY

)()( 22112211 YbYbXaXa

Page 35: Bivariate Linear Correlation. Linear Function Y = a + bX

Uses of Correlation Analysis

• Canonical correlation/regression

• (homophobia, homo-aggression) = (psychopathic deviance, masculinity, hypomania, clinical defensiveness)

• High homonegativity = hypomanic, unusually frank, stereotypically masculine, psychopathically deviant (antisocial)

)()( 22112211 YbYbXaXa

Page 36: Bivariate Linear Correlation. Linear Function Y = a + bX

Factors Affecting Size of r

• Range restrictions– Without variance there can’t be covariance

• Extraneous variance– The more things affecting Y (other then X),

the smaller the r.• Interactions – the relationship between X

and Y is modified by Z– If not included in the model, reduces the r.

Page 37: Bivariate Linear Correlation. Linear Function Y = a + bX

Power Analysis

1 n

Page 38: Bivariate Linear Correlation. Linear Function Y = a + bX

Cohen’s Guidelines

• .10 – small but not trivial• .30 – medium• .50 – large

Page 39: Bivariate Linear Correlation. Linear Function Y = a + bX

PSYC 6430 Addendum

• The remaining slides cover material I do not typically cover in the undergraduate course.

Page 40: Bivariate Linear Correlation. Linear Function Y = a + bX

Correcting for Measurement Error

• If reliability is not 1, the r will underestimate the correlation between the latent variables.

• We can estimate the correlation between the true scores this way:

• rxx and rYY are reliabilities

yyXX

XYYX

rr

rr

tt

Page 41: Bivariate Linear Correlation. Linear Function Y = a + bX

Example

• r between misanthropy and support for animal rights = .36 among persons with an idealistic ethical ideology

.42.)93(.78.

36.

ttYXr

Page 42: Bivariate Linear Correlation. Linear Function Y = a + bX

H: 1 = 2• Is the correlation between X and Y the

same in one population as in another?• The correlation between misanthropy and

support for animal rights was significantly greater in nonidealists (r = .36) than in idealists (r = .02)

Page 43: Bivariate Linear Correlation. Linear Function Y = a + bX

H: WX = WY• We have data on three variables. Does

the correlation between X and W differ from that between Y and W.

• W is GPA, X is SATverbal, Y is SATmath.

• See Williams’ procedure in our text.• See other procedures referenced in my

handout.

Page 44: Bivariate Linear Correlation. Linear Function Y = a + bX

H: WX = YZ

• Raghunathan, T. E, Rosenthal, R, & and Rubin, D. B. (1996). Comparing correlated but nonoverlapping correlations, Psychological Methods, 1, 178-183.

• Example: is the correlation between verbal aptitiude and math aptitude the same at 10 years of age as at twenty years of age (longitudional data)

Page 45: Bivariate Linear Correlation. Linear Function Y = a + bX

H: = nonzero value

• A meta-analysis shows that the correlation between X and Y averages .39.

• You suspect it is not .39 in the population in which you are interested.

• H: = .39.