Week Eight - Linear Regression

Question 1

What type of variables are needed for correlation?

Answer 1

continuous

Question 2

What does r vary between?

Answer 2

-1 and 1

Question 3

What two pieces of information do you get from a correlation coefficient?

Answer 3

The direction of the relationship (positive or negative) and the magnitude of the relationship (weak = 0 or close, strong = closer to -1 or +1)

Question 4

What does a positive relationship look like?

Answer 4

When one variable increases and the other also increases

Question 5

What does a negative relationship look like?

Answer 5

When one variable decreases and the other increases

Question 6

What does the regression line help do?

Answer 6

Predict scores

Question 7

What is the y intercept (a)?

Answer 7

the intercept- the value that we predict the outcome variable to be if their score is 0

Question 8

What is the equation of a line?

Answer 8

Y = b(x) + a

Question 9

What does b represent in the equation?

Answer 9

The slope - how much change in the outcome for each increase in the predictor

Question 10

What is the IV called in linear regression?

Answer 10

Predictor variable

Question 11

What is the DV called in linear regression?

Answer 11

The response or outcome variable

Question 12

What does r give you?

Answer 12

The correlation between the IV and DV

Question 13

What does R squared tell you?

Answer 13

The proportion of variance for a DV that explained by an IV

Question 14

What does adjusted R squared tell you?

Answer 14

The same as R squared but takes into account smaller sample sizes

Question 15

What line in the model coefficients box NEEDS to be significant?

Answer 15

The slope line - has to be significantly different from 0 (the bottom line)

Question 16

Where does the df come from in regression?

Answer 16

From the omnibus table, the t-test comes from the model coefficients table (and the p)

Question 17

Major difference between correlation and regression?

Answer 17

Correlation gives you the strength of the relationship between variables

Regression builds a model of this relo to allow predictions of other scores

Question 18

What is a polynomial equation used for?

Answer 18

Used to describe curvy data that does not fit with the straight line

Question 19

What is a quadratic relationship?

Answer 19

Where there is one bend in the line

Question 20

What do IVs have to be to use a trend analysis?

Answer 20

Quantitative (measurable)

Question 21

What is trend analysis?

Answer 21

Helps to determine whether a linear or higher-order trend is required to describe the relationship between IV and DV

Question 22

How does trend analysis work?

Answer 22

it determines the most complex trend by taking levels of IV and subtracting 1 (2 levels -1 = linear, 3 levels - 1 = quadratic)