Fill in the blank: The ____ will always be the exposure to the intervention -- with 2 levels or sometimes 3.

• Primary IV

• Dependent Variable (Outcome)

• Covariate: Baseline outcome level

• Covariates identified in prep analyses

To minimize confounding, researchers include the ____ as a covariate, even though it is not interpreted.

• Covariates identified in prep analyses

• Baseline level of the outcome

• Primary IV

• Dependent Variable (Outcome)

Fill in the blank: The ____ is the knowledge, attitude, or behavior we are interested in.  It is ALWAYS the post-intervention measure.  There can only be ONE in each model.

• Covariate: Baseline outcome level

• Covariates identified in prep analyses

• Primary IV

• Dependent Variable (Outcome)

Fill in the blank: You should include as independent variables in your model any characteristic variables that were found to be statistically significantly different between the intervention and comparison groups in your prep analyses. These are referred to as: ____.

• Primary IV

• Dependent Variable (Outcome)

• Covariate: Baseline outcome level

• Covariates identified in prep analyses

In ordinary least squares (OLS) regression, R-squared is about...

• About the whole model. The % variability in DV explained by the IVs.

• Not about the whole model. There is one for each IV in the model, null hypothesis is B=0

• About the whole model. H0: taken together, all IVs explain none of the DV variability

In ordinary least squares (OLS) regression, the t-test is about...

• About the whole model. The % variability in DV explained by the IVs.

• About the whole model. Ho: taken together, all IVs explain none of the DV variability.

• Not about the whole model. There is one for each IV in the model, null hypothesis is B=0.

In ordinary least squares (OLS) regression, the F-statistic is about...

• About the whole model. The % variability in DV explained by the IVs.

• About the whole model. Ho: taken together, all IVs explain none of the DV variability.

• Not about the whole model. There is one for each IV in the model, null hypothesis is B=0

TRUE OR FALSE: The confidence intervals are related to the p-value: the coefficient will not be statistically significant if the confidence interval includes the null (0). 

• TRUE

• FALSE

What should be the dummy variable values for ID 004?

• Brooklyn = 1, Manhattan = 1, Queens =1

• Brooklyn = 1, Manhattan = 0, Queens =1

• Brooklyn = 0, Manhattan = 1, Queens =0

Where does Respondent 006 live?

• Brooklyn

• 
  Manhattan

• Queens