1. The government is concerned about long-term unemployment and is thinking about creating a job training program for unemployed workers to increase the rate at which they find jobs. The program consists of attending classes and having meetings with a job counselor. But the government is unsure about the effectiveness of such programs and would like to have an idea of the treatment effect of such programs. The government decides to design a randomized controlled trial to test the effectiveness of the program.
   1. Identify the "treatment," the study population, and an appropriate outcome variable for this setting.
   2. Without a randomized trial, if participation in the job training program were voluntary, how might a comparison of job finding rates between workers who do, and do not enroll in the program be subject to selection bias?
   3. Explain how a randomized experiment solves the problem of selection bias.