13. Assessing Relationships 2

Question 1

measure (or questionnaire) consistently reflects the construct it’s measuring

Answer 1

reliability

Question 2

most common measure of reliability

Answer 2

Chronbach’s alpha

useful with questionnaires

Question 3

Chronbach’s alpha scores

Answer 3

should be .7 to .8 for test retest reliability

Question 4

ICC

Answer 4

measures relationship between 2+ variables that measure the same thing

Question 5

Measurements ICC can be used with

Answer 5

• single measurement

- mean of several measurement

Question 6

ICC can be used to assess:

Answer 6

• inter-rater reliability

- intra-rater reliability

Question 7

3 models of ICC: model 1

Answer 7

each subject assessed by different set of raters

Question 8

3 models of ICC: model 2

Answer 8

• each subject assessed by same set of raters

- raters are representative of the population

Question 9

3 models of ICC: model 3

Answer 9

• each subject assessed by same set of raters

- raters only reliable for their own study

Question 10

Which ICC model is least common?

Answer 10

model 1

Question 11

When is regression used?

Answer 11

when you’re trying to predict an outcome (or DV)

Question 12

variables used to predict the outcome

Answer 12

predictor variables (or IV)

Question 13

Where do predictor variables come from?

Answer 13

previous research

Question 14

types of regression

Answer 14

• simple linear
• multiple linear
• logistic

Question 15

1 IV

1 categorical DV

Answer 15

logistic regression

Question 16

1 IV

1 continuous DV

Answer 16

simple linear regression

Question 17

> 1 IV

1 continuous DV

Answer 17

multiple linear regression

Question 18

data for regression

Answer 18

linear

Question 19

assumptions of a regression: variables

Answer 19

• must be continuous or categorical (with only 2 categories)

- DV is continuous

Question 20

assumptions of a regression: no perfect multicollinarity

Answer 20

two or more predictor variables shouldn’t be highly correlated (i.e. IVs should be measuring different things)

Question 21

multicollinarity and simple linear regression

Answer 21

not a problem

Question 22

assumptions of a regression: confounding variables

Answer 22

predictors are uncorrelated with confounding variables

Question 23

assumptions of a regression: homoscedasticity

Answer 23

• similar to homogeneity of variance

- variance should be equal among predictor variables

Question 24

assumptions of a regression: independence

Answer 24

• values of outcome variable are independent

- come from a separate entry

Question 25

assumptions of a regression: linearity

Answer 25

relationship of IVs and DV is linear

Question 26

What are the assumptions of a regression?

Answer 26

• predictor variables continuous or categorical and DV is continuous
• no perfect multicollinarity
• no confounding variables
• homoscedasticity
• independence
• linearity

Question 27

b_1

Answer 27

slope of the line

Question 28

b_0

Answer 28

y-intercept

Question 29

What are b_1 and b_0?

Answer 29

regression coefficients

Question 30

steps to performing a regression analysis

Answer 30

• assess the model as a whole

- assess individual predictor variables

Question 31

How to assess the model as a whole

Answer 31

goodness of fit