Reliability: The data that we collect is not useful unless we know that it is reliable and valid. Think of the judges on the TV show Dancing with the Stars. (There are three or four judges. Each week each judge scores each dance on a scale from 1 to 10. The dances are different types of dances each week -- waltz, foxtrot, etc).
What data would we have to collect to be sure that the judges ratings meet the criteria for each type of reliability? Be sure to clarify how each type of reliability is different from the other types.