1) To estimate if halfway houses improve the transition from prison, 50 inmates who will be released in a month are randomly assigned to one of two groups. One group is sent to a halfway house to participate in a program called "ReEntry." The other group stays in prison until the time of release. Two months after release, data measuring "success in job placement" are gathered on the subjects in both groups.

a) Use the R, X, O notation to sketch the research design.

b) Someone says that the design does not control for statistical regression since the subjects are inmates and thus represent an "extreme in their social behavior." Do you agree that the design has problems with statistical regression? describe.

c) Define the terms internal validity and external validity in the context of this study.