Week 1

Question 1

Prior knowledge

Answer 1

Bayesian approach. “already known” information.

Question 2

Posterior distribution

Answer 2

Prior knowledge is updates with the information in the data

Question 3

Advantages & disadvantages prior knowledge

Answer 3

Pro: accumulating knowledge & more powered
:( : results depend on the choice of prior

Question 4

Frequentist Pr(data|H0)

Answer 4

P-value. Probability of observing same or more extreme data given that the null is true

Question 5

Bayesian Pr (Hj|data)

Answer 5

Probability that hypothesis Hj is supported by the data

Question 6

Frequentist probability

Answer 6

Relative frequency

Question 7

Bayesian probability

Answer 7

Degree of belief

Question 8

95% confidence interval (frequentist)

Answer 8

If we were to repeat the experiment many times and calculate CI each time, 95% of the intervals will include the true parameter value

Question 9

95% credible interval (bayesian)

Answer 9

There is 95% probability that the true value is in the credible interval

Question 10

R squared

Answer 10

…% of the variance in y is explained by the regression model

Question 11

Adjusted R squared

Answer 11

Corrects for overfitting (having many predictors increase R squared)

Question 12

Method enter (frequentist)

Answer 12

data analist decides what goes in the model (confirmatory)

Question 13

Method stepwise (frequentist)

Answer 13

The best prediction model is determined based on results in this sample (exploratory)

Question 14

B-value

Answer 14

the unstandardized regression B can be used to predict a score on the dependent variable

Question 15

Beta value

Answer 15

the standardized regression coefficient can be used to determine the relative importance of the predictors

Question 16

Registered report

Answer 16

develop idea > design study > stage 1 peer review > collect and analyze data > write report > stage 2 peer review > publish report

Question 17

Cook’s distance

Answer 17

Assumption of no outliers on y-axis. Cook’s distance indicates the overall influence of a respondent on the model. Value must be below 1

Question 18

Violation of absence of multicollinearity leads to…

Answer 18

Regression coefficients (B) are unreliable
Limits magnitude of R (correlation Y and ^Y)
The importance of individual independent variables can hardly be determined, if at all

Question 19

Tolerance or VIF

Answer 19

Determining if multicollinearity is an issue:
Tolerance <.2: indicates potential problem
Tolerance <.1: indicates problem
VIF >10: indicates problem

Question 20

Value of multiple correlation coefficient is same as…

Answer 20

R