Chapter 12 - Correlation Coefficients

Question 1

correlation coefficient

(definition)

Answer 1

A statistic that describes how strongly variables are related to one another

Question 2

What is the Pearson product-moment correlation coefficient?

Answer 2

• The Pearson r
• Values range from 0.00 to ± 1.00

Question 3

On what two scales is the Pearson r used?

Answer 3

Interval and Ratio

Question 4

What two properties of the variable relationship does the Pearson r provide?

Answer 4

Strength and direction

• 0.00 means no r.ship, 1.00 is perfect r.ship
• Closer to 1.00 (plus or minus), stronger r.ship
• r value of ± indicates positive or neg r.ship

Question 5

Which r value represents a stronger variable relationship?
-.54 or +.45

Answer 5

-.54

Question 6

What is a scatterplot used for? (two things)

Answer 6

• To visually show linear patterns of correlation
• Make general prediction about second variable when given the first

Question 7

Regression line
(definition)

Answer 7

In a scatterplot, the line of best prediction (best-fit)

Question 8

What is prediction error?

Answer 8

The difference between actual and predicted (mean) value

Question 9

Regression line equation

Answer 9

Y = bX + a

• (where Y is the score we want to predict, X is the known score, a is a constant, and b is the slope)
• b is the “weighting adjustment factor” [from textbook]

Question 10

What is the purpose of a regression line?

Answer 10

Minimizes distance between actual scores and predicted scores (error)

Question 11

Coefficient of determination, r²
(definition)

Answer 11

• The proportion of variance in an outcome Y that is “explained” or accounted for by the variable
• “Percent of shared variance between two variables” [from textbook]

Question 12

Coefficient of determination, r²
(function)

Answer 12

• Tells us how much is still unknown in our predictions
• Calculated by squaring correlation coefficient r
• Range of 0.00 to 1.00, also can be written as a 0% - 100%

Question 13

What is the difference between r² and R² ?

Answer 13

• r² is used for two-variable regression
• R² is for multiple regression