Regression

Question 1

What is Regression

Answer 1

Predicting 1 variable from another. Involves an IV and a DV

Question 2

What is the general form of a perfect linear relationship

Answer 2

Y = a + bX

Y = Outcome variable
a = Intercept
b = Slope
X = Predictor variable

(usually use Y hat)

Question 3

What is the regression model

Answer 3

Y = a + bX + error

Question 4

What are assumptions of the error or residual term

Answer 4

1. ) Independent
2. ) Homoscedastic (Equal error variance for levels of predicted Y)
3. ) Normally distributed with a mean of 0

Question 5

What is R^2

Answer 5

Proportion of variability in Y that is predicted by X

Question 6

How do we test the reliability of the regression line

Answer 6

Using a F-Test

Question 7

What does the F-Test tell us and what is the procedure

Answer 7

It tell us whether the variance explained is significantly difference from 0. If it is not significant, the regression is worthless (i.e. the predictor X does not explain the outcome Y at all)

Null: R^2 = 0 and use the f -test

Question 8

What is the standardized regression coefficient

Answer 8

Converting of X and Y to Z-scores

Question 9

Significant testing for regression coefficient b

Answer 9

Does the b help us predict the dependent variable Y (i.e. is the slope non-zero)

Null: b = 0 and use the t-test

Question 10

Is the standarized regression coefficient beta same as the correlation coefficient

Answer 10

Yes

Question 11

How do I write up the following: ANOVA and T-Test

A simple regression predicting positive mood from happiness was conducted. 
b=.67
df = 88
R2 = 0.223
F = 25.3
p

Answer 11

Happiness was a significant predictor of mood (b=.67, t(88)=5.03, p