1. In randomization typically researchers report the differences in means of observed characteristics twice once before treatment, and once after treatment. Why do they do that?

a. Before treatment to assess parallel trends, after treatment to estimate the treatment effect on the treated (ATT)

b. Before treatment to ensure external validity, after treatment to ensure internal validity

c. Before treatment to show "balance", after treatment to estimate the treatment effect on the treated (ATT).

d. Before treatment to ensure internal validity, after treatment to ensure external validity

2. For questions a- e please indicate if the statement is True or False and explain?

a. Random assignment always ensures external validity (True or false)

b. With random assignment one can obtain treatment effects by naive with-and-without comparisons.

c. With random assignment, if the difference in outcomes between treatment and control groups before intervention is zero, then the treatment effect is zero.

d. If the treatment effect is zero, then the randomization did not work properly.

e. Regression always measures the causal effect of the independent variable on the dependent

3. Under what condition would you have used a regression discontinuity design (RDD) to evaluate this skills training intervention. (2 MARKS