Lecture 10

Question 1

Linear regression is the next step after finding a sig correlation (pearsons r). What does this tell you ?

Answer 1

How much effect variable X has on variable Y

Question 2

X is which variable?

Answer 2

Independent variable (predictor variable)

Question 3

Y is what type of variable

Answer 3

Dependent variable (criterion variable)

Question 4

In regression analysis the X (independent variable) is called

Answer 4

Predictor

Question 5

Dependent variable is called

Answer 5

Criterion

Question 6

4 types of regression analysis

Answer 6

Simple linear (most common)
Multiple linear
Logistic
Multinominal logistic

Question 7

Simple linear is called simple because

Answer 7

It only has one independent variable (predictor ) and 1 dependent (criterion)

Question 8

In simple linear regression both variables need to be

Answer 8

Numeric - using at least interval scale

Question 9

Simple linear regression allows us to predict the value

Answer 9

That can be assumed by criterion (DV) (Y) if the value of predictor (IV) is known

Question 10

‘Simple’ means

Answer 10

Enter 1 predictor variable at a time to the relationship

Question 11

‘Linear’ implies

Answer 11

There is a constant increasing/decreasing relationship between variables. (Monotonic relationship)

Question 12

What method does SPSS use to calculate simple linear regression?

Answer 12

Ordinary least squares regression OLS

Question 13

In a graph the X (IV) is always on the

Answer 13

Horizontal axis

Question 14

Line if best fit should minimise the dif between the

Answer 14

Actual value of the DV and the value of the DV predicted by the IV

Question 15

Residual is what we call the difference between

Answer 15

The distance between a persons actual score is different (higher or lower that what was predicted for what score they should have gotten (as found on the line of best fit)

Question 16

Residual represents the

Answer 16

Error in your model

Question 17

The smaller the residual the smaller

Answer 17

Your error

Question 18

Big residuals indicate that

Answer 18

Your model chosen is not a good fit for the data

Question 19

Some residuals are negative which means that the predicted score is

Answer 19

Over estimation

Question 20

Some residuals are positive which means the predicted score is

Answer 20

Under estimation

Question 21

What does the ANOVA tell you

Answer 21

Goodness of fit

Question 22

ANOVA needs to be ………..to go further and assume goodness of fit

Answer 22

Sig

Question 23

Correlation only allows you to identify

Answer 23

A relationship between X and Y