Correlation

Question 1

What is correlation?

Answer 1

The extent to which two variables are linearly related

Question 2

1st stage of correlation

Answer 2

See if they covary an if variance in one predicts variance in the other

Question 3

Pearson’s r calculation

Answer 3

Cov(x,y) = sum of (xi - xmean) (yi - ymean) / n - 1

Question 4

Effect of r

Answer 4

.1 = small, .3 = medium, .5 = large

Question 5

Positive line (correlation)

Answer 5

Points up

Question 6

Negative line (correlation)

Answer 6

Points down

Question 7

On a graph of correlation

Answer 7

x axis is independent, y axis is outcome

Question 8

What does the slope on a correlation graph mean?

Answer 8

Strength of correlation

Question 9

Co-efficient of determintion

Answer 9

r2 provides an equal interval and ratio measure of co-variation

Question 10

Assumptions for correlation

Answer 10

Data is continuous, independent, linear, normal distribution

Question 11

If assumptions are violated (correlation)

Answer 11

Use Spearman rho, Kendall’s tau or bootstrapping

Question 12

Biserial correlation

Answer 12

When one variable is dichotomous but on an underlying continumn

Question 13

Point-biserial correlation

Answer 13

When one variable is truly dichotomous

Question 14

What type of correlation is most commonly used

Answer 14

Bivariate correlation

Question 15

Partial correlation

Answer 15

Can hold a 3rd variable constant and see true relationship between 1 and 2. It removes the 3rd variables contribution to predictor and outcome

Question 16

Semi-partial correlation

Answer 16

Removes 3rd variable contribution to predictor but not outcome